- Email: [email protected]
Co-localized but target-unrelated expression of vasoactive intestinal polypeptide and galanin in rat dorsal root ganglion neurons after peripheral nerve crush injury
Brain Research, 582 (1992) 47-57 © 1992 Elsevier Science Publishers B.V. All rights reserved. 0006-8993/92/$05.00
47
BRES 17775
Co-localized but target-unrelated expression of vasoactive intestinal polypeptide and galanin in rat dorsal root ganglion neurons after peripheral nerve crush injury Hitoshi Kashiba a, Emiko Senba b'c, Yoshihiro Ueda a and Masaya Tohyama b aDepartment of Physiology, Kansai College of Acupuncture Medicine, 990 Ogaito, Kumatori, Sennan, Osaka, 590-04 (Japan), bDepartment of Anatomy (II), Osaka University, Medical School, 2-2 Yamadaoka, Suita, Osaka, 565 (Japan) and CDepartment of Anatomy (I1), Wakayama Medical College, 27 9 Ban cho, Wakayama, 640 (Japan) (Accepted 21 January 1992) Key words: Vasoactive intestinal polypeptide; Galanin; Neurofilament; Primary sensory neuron; Nerve crush; Retrograde-tracing
Expression of vasoactive intestinal polypeptide (VIP) and galanin in dorsal root ganglion (DRG) neurons is known to be induced by peripheral nerve injury. We investigated (1) whether VIP and galanin were co-expressed by DRG neurons and (2) whether such neurons innervated specified peripheral targets (visceral, cutaneous or muscular). An antibody to the 200 kDa neurofilament subunit (NF200) was used as a marker for large type-A cells in the DRG. VIP and galanin were respectively observed in 22% and 67% of DRG neurons at the L5 spinal level after crushing of the sciatic nerve. Most VIP-containing neurons were small type-B ceils (about 90%) and approximately 95% of VIP-containing neurons also showed galanin-like immunoreactivity. Galanin was expressed by both large type-A and small type-B cells. Immunocytochemistry combined with a retrograde tracer revealed that about 70-80% of the small type-B cells in each sensory division displayed VIP-like immunoreactivity, and that most of the tracer-labeled neurons also expressed galanin. These findings suggest that the expression VIP and/or galanin in response to peripheral nerve crush injury is a property common to visceral, cutaneous and muscular sensory neurons. INTRODUCTION Various neuropeptides have been discovered in rat dorsal root ganglion ( D R G ) cells 11. It is known that the peptide and m R N A levels of vasoactive intestinal peptide (VIP), the VIP-related peptide with amino-terminal histidine and carboxy-terminal isoleucine (PHI, a peptide which is coded on V I P precursor m R N A ) , and galanin increase following nerve crush or transection 8'19'25'26' 30,32. In contrast, these peptides (especially VIP) are present at very low levels in normal D R G neurons. The newly VIP-containing neurons are exclusively small D R G neurons 3°, while galanin-containing neurons are both small and large 8. It is reported that some D R G neurons with injured axons react for both V I P / P H I and galanin antisera 36, but in this case the proportion of coexistence is not known. In the present study, we further investigated the proportions of VIP- and/or galanin-positive neurons and their cell type (large type-A or small type-B) in animals with sciatic nerve crush injuries. It is well documented that D R G neurons contain various combinations of neuropeptides, and that the chemicals expressed are specified by their peripheral targets.
D R G neurons can be divided into three groups acording to their peripheral targets, i.e. those with cutaneous, muscular and visceral target tissues. These peripheral tissues originate from different dermal layers during the developmental process. The chemical expression of D R G neurons which innervate skin and viscera 12'13'22 or skin and muscles 28 have been demonstrated to be different. Also, when their peripheral processes are connected to new locations, peptide expression is altered to become appropriate to the new target 2°. It is thus reasonable to assume that neuronal peptide expression is regulated by target-derived neurotrophic factors (NTFs). Moreover, each neuropeptide seems to be regulated by a specific mechanism involving NTFs. For example, the production of substance P (SP) and somatostatin (SOM) by cultured D R G neurons is regulated by different NTFs 23. Expression of SP and calcitonin gene-related peptide ( C G R P ) in D R G neurons is up-regulated by exogenously administered nerve growth factor (NGF) both in vivo 7 and in vitro 17, while the expression of V I P and galanin seems to be independent of N G F 14'24. We showed in our previous study that blockade of axonal flow at a peripheral site causes an increase in the level
Correspondence: H. Kashiba, Department of Physiology, Kansai College of Acupuncture Medicine, 990 Ogaito, Kumatori, Sennan, Osaka, 590-04, Japan.
48 of these peptides in D R G cells in vivo, suggesting that synthesis of these two peptides in negatively controlled by target-derived NTFs under normal conditions TM. Various kinds of NTFs are tissue specific, such as brain-derived neuro-trophic factor (BDNF) 2 and motor neuron growth factor (MNGF) 2~. Accordingly, the question may be raised as to whether the expression of VIP and galanin is target-related or not. The second aim of this study was to investigate
whether the expression of these peptides in D R G neurons was differentially controlled by their peripheral targets or not. Neurons were labeled by injecting retrograde tracers into the respective peripheral nerve branches innervating skin, muscle and visceral tissues. Subsequently, we examined which neuron groups contained VIP and/or galanin. The distributions of central processes containing newly synthesized V I P or galanin in the dorsal horn was also investigated. A n antibody to the
Fig. 1. Pairs of fluorescent photomicrographs showing VIP-IR (A,C), galanin-IR (B,E) and NF200-IR neurons (D,F) in adjacent serial sections (A and B) and in the same section (C and D, E and F) of the DRG (L5) after crushing of the sciatic nerve. Double-labeled ceils are indicated with arrowheads. Bar = 100 ~m.
49 200 k D a s u b u n i t o f n e u r o f i l a m e n t s ( N F 2 0 0 ) w a s u s e d in t h i s s t u d y as a m a r k e r f o r l a r g e t y p e - A D R G
cells 16'35.
MATERIALS AND METHODS
Tissue preparations Male Wistar rats weighing about 200 g were used. All the animals were divided into three groups and were operated on under sodium pentobarbital anesthesia (50 mg/kg). Those in the first group (n = 4) received crushing of the unilateral sciatic nerve at the hip joint level by tapping it with a pair of forceps for 30 s (about 60 times). The animals in the second group were given injections of a 2% (w/v) solution of fluoro-gold (FG) in distilled water into the unilateral greater splanchnic nerve (0.2/~l) (n = 3), the unilateral sural nerve (the major cutaneous branch of the tibial nerve) (0.4/A) (n = 3), or the unilateral muscular branches of the tibial nerve (0.2 #1 per branch) (n = 3). Injection was performed using a glass micropipette with a tip diameter of 20-50/~m connected to a microinjector. The peripheral regions of the injected nerves were then crushed by tapping with forceps as described above. As a control experiment, animals in the third group (n = 2) received FG injection into the sural nerve or the greater splanchnic nerve and the nerves proximal to the injection site were ligated and cut. All the animals in the three groups were kept alive for 7 days after these treatments. All rats were perfused under deep sodium pentobarbital anesthesia (75 mg/kg) with 0.9% saline (100 ml) followed by cold Zamboni's fixative (400 ml) 37. The spinal cords with the bilateral DRGs corresponding to the spinal levels of the treated nerves (greater splanchnic nerve; T9, sural nerve, muscular branches of the tibial nerve, and sciatic nerve; L5) were excised, post~fixed in the same fixative overnight, and placed in 30% (w/v) sucrose in 0.1 M phosphate buffer for 2 days at 4°C. Spinal cords and DRGs were then cut into sections on a cryostat at -20°C, and collected in small wells or mounted on gelatin-coated glass slides.
(Amersham; 1:250) or TR-conjugated sheep anti-mouse Ig (1:250). The preparations were photographed under a fluorescence microscope with U and G filters so as to visualize only FG and TR fluorescence, respectively. Serial sections of DRGs were cut, and the immunoreactive or FG-labeled cells in every fifth section were counted on enlarged photographic prints. The shortest diameters of immunoreactive neurons labeled by FG were also measured. Neurons in the D R G were divided into four groups according to their size; small (<20/~m in diameter), intermediate (20-30/~m), large (30-45/zm), and giant (>45/~m). In group 3 (control), slidemounted D R G sections were air-dried, coverstipped and examined.
Avidin-biotin-peroxidase complex (ABC) method Immunocytochemical demonstration of VIP in the spinal cord was performed by the method of Hsu et al. 9 using an ABC kit (Vector). After incubation, sections were visualized by reaction with 0.02% (w/v) 3.3'-diaminobenzidine (DAB) and 0.005% (v/v) H20 2 in Tris-HCl buffer for 10-30 min. Then the sections were mounted on gelatin-coated slides, dehydrated, and coverslipped with Permount (Fisher Scientific Co.).
Specificity of the antisera The polyclonal antibodies to galanin and VIP were produced against purified natural porcine galanin and VIP, respectively. The specificity of each of these antibodies was checked by absorption tests. VIP-like immunoreactive (IR) or galanin-IR structures were not stained in sections pre-absorbed with synthetic galanin and VIP at 104 M. The monoclonal antibody to NF200 was raised against 200 kDa porcine NE This antibody has been shown to specifically recognize the 200 kDa subunit of NF in a number of mammalian and avian species 29.
RESULTS
Double-staining o f VIP, galanin and NF200 Double-staining for VIP, galanin and NF200 The indirect immunofinorescent method of Coons 5 was used. For double-staining of VIP and galanin in group 1, adjacent 5-/~m serial sections of the L5 D R G were collected onto separate slides, and were incubated for 2 days at 4°C with rabbit anti-VIP (Ortho Diagnostic Systems; 1:25) and rabbit anti-galanin (Serotec; 1:2,000) antiserum diluted with 0.02 M PBS containing 0.3% Triton X. The sections were then washed in 0.02 M PBS and then incubated overnight at 4°C with fluorescein isothiocyanate (FITC)-conjugated sheep anti-rabbit IgG (Miles; 1:1,000). After a final wash, the samples were air-dried and coverslipped with a glycerine and PBS mixture (1:1). The preparations were viewed and photographed under a fluorescence microscope equipped with a B excitation filter. For double-staining of NF200 and VIP, or NF200 and galanin in group 1, slide-mounted 10-/tm D R G sections were first incubated in a mixture of mouse anti-NF200 monoclonal antibody (ICN Biochemicals, Inc.; 1:10,000) and rabbit anti-VIP (1:25) or rabbit antigalanin (1:2,000) polyclonal antibody.. Samples were then incubated with a mixture of Texas red (TR)-conjugated sheep anti-mouse Ig (Amersham; 1:250) and FITC-conjugated sheep anti-rabbit IgG (1:1,000). Preparations were viewed and photographed under an Olympus fluorescence microscope equipped with G and B excitation filters to assure exclusive visualization of TR and FITC fluorescence, respectively. Preliminary control experiments showed no cross-reactions between the primary and secondary antibodies.
Combined retrograde fluorescent dye tracing and immunofluorescence for V1P~ galanin and NF200 For the immunocytochemical demonstration of VIE galanin or NF200 in group 2, slide-mounted sections (10-/tm thick) or freely floating sections (25-/~m thick) were incubated with anti-VIP, antigalanin or anti-NF200 antiserum as described above. The samples were then incubated with TR-conjugated donkey anti-rabbit Ig
VIP-like immunoreactivity (IR) and galanin-IR were r a r e in t h e c o n t r o l D R G s o n t h e c o n t r a l a t e r a l side ( a b o u t 1 and 3% of the DRG neurons, respectively). These neur o n s w e r e all s m a l l t o i n t e r m e d i a t e i n size. O n t h e o t h e r hand,
many VIP-IR
or galanin-IR neurons
were ob-
s e r v e d in t h e i p s i l a t e r a l D R G s a f t e r n e r v e c r u s h i n g (Figs. 1, 2 a n d 3). T h e n u m b e r
of galanin-IR neurons was
TABLE I
Incidences of VIP-1R, galanin-lR and NF2OO-IR in the visceral, cutaneous and muscular DRG neurons of axotomized rats
VIP
mean_+ S.E.M. Galanin
mean + S.E.M. NF200
mean + S.E.M.
Visceral (n = 3)
Cutaneous (n = 3)
Muscular (n = 3)
58/89 (65%) 55/73 (75%) 29/52 (56%) 6 5 _ 5% 76/82 (93%) 48/54 (89%) 32/35 (91%) 91 + 1% 5/68 ( 7 % ) 6/73 ( 8 % ) 8/69 (12%) 9 + 2%
116/251 (46%) 44/87 (51%) 79/153 (52%) 50 + 2% 104/120 (87%) 61/69 (88%) 66/80 (83%) 86 + 2% 31/109 (28%) 45/139 (32%) 37/101 (37%) 32 + 3%
7/25 (28%) 11/42 (26%) 7/45 (16%) 23 + 4% 21/28 (75%) 31/37 (84%) 27/40 (68%) 76 + 5% 27/37 (73%) 45/71 (63%) 29/37 (78%) 71 + 4%
50
Fig. 2. Pairs of fluorescent photomicrographs showing FG-labeled visceral (A), cutaneous (C) and muscular (E) sensory cells and VIP-IR neurons (B,D,F) in the same sections (A and B, C and D, E and F) of the DRG after nerve crush injury. FG-labeled immunoreactive cells are indicated by arrowheads. Bar = 100/~m.
much greater than that of V I P - I R neurons, with 22 and 67% of the D R G neurons (L5) being positive for VIP and galanin, respectively. The results of double-staining for V I P and galanin are shown in Fig. 1. Ninety-four percent of the V I P - I R neurons (59/63 cells) also displayed galanin-IR and 29% of the galanin-IR neurons (59/202 cells) were positive for VIP. NF200-IR neurons and V I P - I R or galanin-IR neurons in the L5 D R G ipsilateral to the crushed sciatic nerve
are shown in Fig. 1. A b o u t 35% of both the ipsilateral injured and contralateral normal D R G neurons showed NF200-IR. These neurons were intermediate to giant in size. Neurons intensely positive for NF200 were often detected in the ipsilateral D R G s . Some V I P - I R or galanin-IR neurons were also positive for NF200 (Fig. 1), with 9% (12/131 cells) and 43% (43/100 cells) of the NF200-IR neurons positive for V I P and galanin, respectively. Conversely, NF200 was detected in 9% (10/110
51
Fig. 3. Pairs of fluorescent photomicrographs showing FG-labeled visceral (A), cutaneous (C) and muscular (E) sensory cells and galanin-IR cells (B,D,F) in the same sections (A and B, C and D, E and F) of the DRG after nerve crush injury. FG-labeled immunoreactive cells are indicated by arrowheads. Bar = 100/tm.
cells) of V I P - I R neurons and 25% (43/173 cells) of galan i n - I R neurons, respectively.
FG and VIP-IR, galanin-IR, or NF2OO-IR V I P - I R , galanin-IR, and NF200-IR neurons as well as F G - l a b e l e d visceral, cutaneous and muscular afferent neurons are shown in Figs. 2, 3 and 4, respectively. A quantitative evaluation of these neurons is shown in Table I. The distribution of the diameters of the immuno-
reactive neurons labeled with F G fluorescence in Fig. 5, together with the m e a n diameters ___ S.D. Positivity for VIP, galanin and NF200 was relatively constant within each group of neurons, but differed between the neurons innervating different tissues (Table I). Injection of F G into the sural nerve p r o d u c e d fluorescence in a large n u m b e r of D R G cells of various sizes from small to giant, in contrast to injection of F G into the greater splanchnic nerve and the muscular branches
52
Fig. 4. Pairs of fluorescent photomicrographs showing FG-labeled visceral (A), cutaneous (C) and muscular (E) sensory cells and NF200-IR cells (B,D,F) in the same sections (A and B, C and D, E and F) of the DRG after nerve crush injury. FG-labeled immunoreactive cells are indicated by arrowheads. Bar = 100 ~m.
of the tibial nerve (Figs. 2, 3 and 4). Many large to giant muscular sensory neurons showed F G fluorescence, but none of the labeled visceral sensory neurons were giant. In control experiments (group 3), no F G fluorescent cells were observed in the T9 D R G of the rat receiving transection of the greater splanchnic nerve after F G injection or in the L5 D R G of the rat in which the sural nerve was cut after F G injection, showing that F G
was specifically transported via these neuronal processes. V I P - I R was observed in 65 and 50% of the visceral and cutaneous sensory neurons, respectively. Most of these neurons were small to intermediate in size. Twentythree percent of muscular afferent neurons were positive for V I P and these neurons were mainly intermediate to large in size. Neurons of various sizes with F G fluorescence were mostly positive for galanin. G a l a n i n - I R was
301
20
53
visceral 6596 23. - •
l~r~
Galanin visceral 91~ 23.9 _+ 6.5 Hm
~ ~ _ . m
NF200 visceral
9t 31.2 + 6.6 Hm
10
0 • I~ m t--
20
~
lO
. . . .
cutaneous 86g
cutaneous 50%
~
201 10 0
23tmusclar 33.4 + 10.1 Hm , . O
20
40
6"0
muscular 76~
~ 20 Cell
.
40 diameter
60
~
.
.--i
. . . .
cutaneous 32~ 38.9 + 6.6 ~Jm
24.9 + 7.9 lJm
o
r ~ . .
7
_
p
muscular 71~ 42.1 _
m
0
20
+
~
40
60
(ijm)
Fig. 5. Histograms showing the distribution of the cell diameter of VIP-IR (left), galanin-IR (center) and NF200-IR (right) neurons labeled with FG injected into the greater splanchnic nerve (upper), the sural nerve (middle) and the muscular branches of the tibial nerve (lower), respectively. The left axis shows the percentage of immunoreactive cells among the FG-labeled neurons in each sensory division. Arrows on the abscissa indicate the mean diameters in each sensory division.
observed in 91, 86 and 76% of visceral, muscular sensory neurons, respectively. observed in 9, 32 and 71% of visceral, muscular sensory neurons, respectively, rons were all relatively large (Fig. 5).
cutaneous and NF200-IR was cutaneous and and these neu-
VIP-IR or galanin-IR fibers in the dorsal horn Dorsal horn VIP-IR and galanin-IR fibers on the ipsilateral and contralateral sides are shown in Figs. 6 and 7, respectively. Little VIP-IR was observed in the contralateral dorsal horn at the T9 and L5 spinal levels. On the other hand, galanin-IR was intense in laminae I - I I of the dorsal horn at the same levels. After crushing of the greater splanchnic nerve, both VIP-like and galaninlike immunoreactivities were slightly increased in the medial superficial laminae of the T9 dorsal horn (Figs. 6B and 7B). After crushing of the cutaneous (sural) or muscular branches of the tibial nerve, VIP-like and galanin-like immunoreactivities were increased in the central areas of laminae I - I I (Figs. 6D,F and 7D,F). VIP-IR cutaneous fibers were more widely distributed mediolaterally than VIP-IR muscular fibers. A small number of galanin-IR fibers were also observed in the deeper layer (laminae III), where no VIP-IR fibers were found. No apparent changes of both kinds of immunoreactivity were observed in other parts of the dorsal horn. DISCUSSION
The number of galanin-IR neurons appearing after nerve crush was about three times that of VIP-IR neu-
rons. A quarter of the galanin-IR neurons also showed NF200-IR while only about 10% of VIP-IR neurons were positive for NF200, indicating that most VIP-IR neurons may be small type-B cells (which are negative for NF200). Co-expression of VIP and galanin was demonstrated in injured sensory neurons, which confirmed previous observations 36. We further demonstrated that about 95% of VIP-positive neurons also displayed ga!anin-IR, while two-thirds of galanin-IR neurons were negative for VIP. These findings suggest that VIP and galanin may be expressed together by damaged small type-B cells, while only galanin is expressed by some of the larger neurons when their peripheral processes are injured. However, very little is known about the physiological functions of VIP and galanin in injured neurons. An increase of VIP and galanin was demonstrated in visceral, cutaneous and muscular sensory neurons following nerve crush in the present study, but the proportion of VIP-IR neurons differed between these three sensory divisions. The proportions of NF200-IR neurons also differed between each sensory division, as has been previously reported 13'2s. Most muscular sensory neurons contained NF200 but visceral sensory neurons only rarely expressed this substance. The size distribution of these neurons was also different. These results indicate that the proportions of large type-A and small type-B cells were different between each sensory division. Most VIP-IR neurons were small type-B cells, as mentioned above, and the proportion of VIP-positive neurons among the small cells remained constant at 70-80% throughout the three sensory divisions. Galanin-IR neu-
54
~f
Fig. 6. Bright-field photomicrographs showing VIP-IR fibers in the dorsal horn at spinal levels corresponding to the crushed nerves; greater splanchnic nerve (upper), sural nerve (middle), and muscular branches of the tibial nerve (lower). The regions where changes in immunostaining were observed on the ipsilateral side (B,D,F) compared to the contralateral side (A,C,E) are indicated by arrowheads. Bar = 200
ktm. rons also do not appear to specifically innervate one of these types of target tissues, because most D R G neurons in each sensory division expressed galanin (76-91%). In our previous report TM, we suggested that the synthesis of VIP and galanin is controlled by negative feedback from target-derived NTF(s). The present study would suggest that these factors, if they exist, are not tissue-specific. Newly synthesized V I P and galanin may be centrally
and peripherally transported, respectively, because their immunoreactivities or contents increase in both the dorsal horn and in the injured peripheral nerve proximal to the lesion 1'33. Therefore, there have been several hypotheses about their functional roles in the central and peripheral regions. In addition to its central analgesic action 15, VIP has been suggested to stimulate glycogenolysis TM, act as an NTF for glial cells 3 and act as a
55
Fig. 7. Immunofluorescent photomicrographs showing galanin-IR fibers in the dorsal horn at spinal levels corresponding to the crushed nerves; greater splanchnic nerve (upper), sural nerve (middle), and muscular branches of the tibial nerve (lower). The regions where changes in immunostaining were observed on the ipsilateral side (B,D,F) compared to the contralateral side (A,C,E) are indicated by arrowheads. Bar = 200/~m.
vasodilator 34. It may contribute to the process of neuronal regeneration through these diverse functions. On the other hand, galanin has been reported to control the release of V I P 1°, dopamine 27, norepinephrine 31 and acetylcholine 6 in the central nervous system, although none of these transmitters apart from V I P has been identified in D R G neurons. Galanin may regulate the release of VIP from the central branches of the primary sensory
neurons to control the above-mentioned actions of VIP during the process of axonal regeneration. Inhibitory effects of intrathecally administered galanin on VIP-induced facilitation of spinal flexor reflex has been demonstrated in axotomized rats 36. In this context, it is of great interest to show the distribution of central branches containing V I P and galanin in axotomized rats. Sites of projection were quite simi-
56 lar between VIP-IR and galanin-IR fibers, in spite of differences of cell size and type of VIP-IR and galanin-IR neurons in the DRG. It is surprising that these peptide-containing fibers were apparently confined to the laminae I and II. A close association between the functional type of peripheral receptors and area of their afferent terminals within the dorsal horn has been well documented (for review see ref. 4); large afferent fibers originated from type-A neurons connected to low threshold mechanoreceptors in the skin or receptors in intrafusal muscles terminate in the deeper laminae, while A6 and C fibers connected to nociceptors or thermoreceptors are distributed exclusively in laminae I and Iio
where nociceptive-specific dorsal horn neurons are localized. Most of the newly appeared galanin-IR fibers were confined to laminae I and II, although at least a quarter of galanin-IR neurons were type-A in axotomized DRG cells. These findings, taken together, may indicate that DRG neurons which synthesize VIP and/or galanin are mainly involved in thermo-nociceptive transmission mechanism.
REFERENCES
13 Kashiba, H., Senba, E., Ueda, Y. and Tohyama, M., Cell size and cell type analysis of calcitonin gene-relateed peptide-containing cutaneous and splanchnic sensory neurons in the rat, Peptide, 12 (1991) 101-106. 14 Kashiba, H., Senba, E., Kawai, Y., Ueda, Y. and Tohyama, M., Axonal blockade induces the expression of vasoactive intestinal polypeptide and galanin in the rat dorsal root ganglion neurons, Brain Res., in press. 15 Komisaruk, B.R., Banas, C., Mehta, A., Cash, P., Whipple, B. and Jordan, E, Analgesic effect of synthetic fragments of vasoactive intestinal peptide in the rat, Soc. Neurosci. Abst., 16 (1990) 565. 16 Lawson, S.N., Harper, A.A., Harper, E.I., Garson, J.A. and Anderton, B.H., A monoclonal antibody against neurofilament protein specifically labels a subpopulation of rat sensory neurons, J. Comp. Neurol., 228 (1984) 263-272. 17 Lindsay, R.M. and Harmar, A.J., Nerve growth factor regulates expression of neuropeptide genes in adult sensory neurons, Science, 337 (1989) 362-364. 18 Magistretti, P.J., Morrison, J.H., Shoemaker, W.J., Sapin, V. and Bloom, EE., Vasoactive intestinal polypeptide induces glycogenolysis in mouse cortical slices: a possible regulatory mechanism for the local control of energy metabolism, Proc. Natl. Acad. Sci. USA, 78 (1981) 6535-6539. 19 McGregor, G.P., Gibson, S.J., Sabate, I.M., Blank, M.A., Christofides, N.D., Wall, P.D., Polak, L.M. and Bloom, S.R., Effect of peripheral nerve section and nerve crush on spinal cord neuropeptides in the rat: increase VIP and PHI in the dorsal horn, Neuroscience, 13 (1984) 207-216. 20 McMahon, S.B. and Gibson, S., Peptide expression is altered when afferent nerves re-innervate inappropriate tissue, Neurosci. Lett., 73 (1987) 9-15. 21 McManaman, J.L., Crawford, EG., Stewart, S.S. and Appel, S.H., Purification of a skeletal muscle polypeptide which stimulates choline acetyltransferase activity in cultured spinal cord neurons, J. Biol. Chem., 263 (1988) 5890-5897. 22 Molander, C., Ygge, J. and Dalsgaard, C.J., Substance P-, somatostatin- and calcitonin gene-related peptide-like immunoreactivity and fluoride resistant acid phosphatase-activity in relation to retrogradely labeled cutaneous, muscular and visceral primary sensory neurons in the rat, Neurosci. Lett., 74 (1987) 37-42. 23 Mudge, A.W., Effect of chemical environment on levels of substance P and somatostatin in cultured sensory neurons, Nature, 292 (1981) 764-767. 24 Mulderry, P.K. and Lindsay, R.M., Rat dorsal root ganglion neurons in culture express vasoactive intestinal polypeptide (VIP) independently of nerve growth factor, Neurosci. Lett., 108 (1990) 314-320. 25 Nielsch, U. and Keen, P., Reciprocal regulation of tachykininand vasoactive intestinal peptide-gene expression in rat sensory
1 Anand, P., Gibson, S.J., Scaravivlli, F., Blank, M.A., McGregor, G.E, Appenzeller, O., Dhital, K., Polack, J.M. and Bloom, S.R., Studies of vasoactive intestinal polypeptide expression in injured peripheral neurons using capsaicin, sympathectomy and m f mutant rats, Neurosci. Lett., 118 (1990) 60-61. 2 Barde, Y.-A., Edgar, D. and Thoenen, H., Purification of a new neurotrophic factor from mammalian brain, EMBO J., 1 (1982) 549-553. 3 Brenneman, D.E., Neale, E.A., Foster, G.A., d'Autremont, S.E. and Westbrook, G.L., Non-neuronal cells mediate neurotrophic action of vasoactive intestinal peptide, J. Cell. Biol., 104 (1987) 1603-1610. 4 Cervero, E, Dorsal horn neurons and their sensory inputs. In T.L. Yaksh (Ed.), Spinal Afferent Processing, Plenum, New York, (1986) pp. 197-216. 5 Coons, A.H., Fluorescent antibody methods. In J.E Danielli (Ed.), General Cytochemical Method, Academic Press, New York, (1958) pp. 399-422. 6 Fiscone, G., Wu, C.E, Consolo, E, Nordstrom, O., Brynne, N., Bartfai, T., Melander, T. and Hokfelt, T., Galanin inhibits acetylcholine release in the ventral hjppocampus of the rat: histochemical autoradiographic in vivo and in vitro studies, Proc. Natl. Acad. Sci. USA, 88 (1987) 7339-7343. 7 Fitzgerald, M., Wall, ED., Goedert, M. and Emson, EC., Nerve growth factor counteracts the neurophysiological and neurochemical effects of chronic nerve section, Brain Res., 332 (1985) 131-141. 8 Hokfelt, T., Wiesenfeld-Hallin, Z., Villar, M.J. and Melander, T., Increase of galanin-like immunoreactivity in rat dorsal root ganglion cells after peripheral axotomy, Neurosci. Lett., 83 (1987) 217-220. 9 Hsu, S.-M., Raine, L. and Fanger, H., Use of avidin-biotinperoxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabelled antibody (PAP) procedures, J. Histochem. Cytochem., 29 (1981) 577-588. 10 Innoue, T., Kato, Y., Koshiyama, H., Yanaihara, N. and Imura, H., Galanin stimulates the release of vasoactive intestinal polypeptide from perfused hypothalamic fragments in vitro and from periventricular structures into the cerebrospinal fluid in vivo in the rat, Neurosci. Lett., 85 (1988) 95-100. 11 Ju, G., Hokfelt, T., Brodin, E., Fahrenkrug, J., Fischer, J.A., Frey, P., Elde, R.P. and Brown, J.C., Primary sensory neurons of the rat showing calcitonin gene-related peptide immunoreactivity and their relation to substance P-, somatostatin-, galanin-, vasoactive intestinal polypeptide- and cholecystokinin-immunoreactive ganglion cells, Cell Tissue Res., 247 (1987) 417-431. 12 Kashiba, H., Senba, E., Ueda, Y., Tohyama, M., Calbindin D28k-containing splanchnic and cutaneous dorsal root ganglion neurons of the rat, Brain Res., 528 (1990) 311-316.
Acknowledgements. This work was partly supported by a Grantin-Aid from the Ministry of Education, Science and Culture of Japan (03770035).
57
26
27
28
29
30
31
neurons following cut and crush injury, Brain Res., 481 (1989) 25-30. Noguchi, K., Senba, E., Morita, Y., Sato, M. and Tohyama, M., Prepro-VIP and preprotachykinin mRNAs in the rat dorsal root ganglion cells following peripheral axotomy, Mol. Brain Res., 6 (1989) 327-330. Nordstrom, O., Melander, T., Hokfelt, T., Bartfal, T. and Goldstein, M., Evidence for an inhibitory effect of the peptide galanin on dopamine release from the rat median eminence, Neurosci. Leg., 73 (1987) 21-26. O'Brien, C., Woolf, C.J., Fitzgerald, M., Lindsay, R.M. and Molander, C., Difference in the chemical expression of rat primary afferent neurons which innervate skin, muscle or joint, Neuroscience, 32 (1989) 493-502. Shaw, G., Debus, E. and Weber, K., The immunological relatedness of neurofilament proteins of higher vertebrates, Eur. J. Cell Biol., 34 (1984) 130-136. Shehab, S.A.S., Atkinson, M.E. and Payne, J.N., The origin of the sciatic nerve and changes in neuropeptides after axotomy: a double-labelling study using retrograde transport of True blue and vasoactive intestinal polypeptide immunohistochemistry, Brain Res., 376 (1986) 180-185. Tsuda, K., Yokoo, H. and Goldstein, M., Neuropeptide Y and galanin in norepinephrine release in hypothalamic slices, Hy-
pertension, 14 (1989) 81-86. 32 Villar, M.J., Cortes, R., Theodorsson, E., Wiesenfeld-Hallin, Z., Schalling, M., Fahrenkrug, J., Emson, P.C. and Hokfelt, T., Neuropeptide expresion in rat dorsal root ganglion cells and spinal cord after peripheral nerve injury with special reference to galanin, Neuroscience, 33 (1989) 587-604. 33 Villar, M.J., Wiesenfeld-Hallin, Z., Xu, X.-J., Theodorsson, E., Emson, P.C. and Hokfelt, T., Further studies in galanin-, substance P- and CGRP-like immunoreactivities in primary sensory neurons and spinal cord: effects of dorsal rhizotomies and sciatic nerve lesions, Exp. Neurology, 112 (1991) 29-39. 34 Williams, T.J., VIP is more potent than PGE 2 as a vasodilator and oedema potentiator in rabbit skin, Br. J. Pharmacol., 77 (1982) 505-509. 35 Wood, J.N. and Anderson, B.H., Monoclonal antibodies to mammalian neurofilaments, Biosci. Rep., 1 (1981) 263-268. 36 Xu, X.-J., Wiesenfeld-Hallin, Z., Villar, M.J., Fahrenkrug, J. and Hokfelt, T., On the role of galanin, substance P and other neuropeptides in primary sensory neurons of rat: studies on spinal reflex excitability and peripheral axotomy, Eur. J. Neurosci., 2 (1990) 733-743. 37 Zamboni, L. and De Martino, C., Buffered picric acid formaldehyde: a new rapid fixative for electron microscopy, J. Cell Biol., 35 (1967) 148A.
47
BRES 17775
Co-localized but target-unrelated expression of vasoactive intestinal polypeptide and galanin in rat dorsal root ganglion neurons after peripheral nerve crush injury Hitoshi Kashiba a, Emiko Senba b'c, Yoshihiro Ueda a and Masaya Tohyama b aDepartment of Physiology, Kansai College of Acupuncture Medicine, 990 Ogaito, Kumatori, Sennan, Osaka, 590-04 (Japan), bDepartment of Anatomy (II), Osaka University, Medical School, 2-2 Yamadaoka, Suita, Osaka, 565 (Japan) and CDepartment of Anatomy (I1), Wakayama Medical College, 27 9 Ban cho, Wakayama, 640 (Japan) (Accepted 21 January 1992) Key words: Vasoactive intestinal polypeptide; Galanin; Neurofilament; Primary sensory neuron; Nerve crush; Retrograde-tracing
Expression of vasoactive intestinal polypeptide (VIP) and galanin in dorsal root ganglion (DRG) neurons is known to be induced by peripheral nerve injury. We investigated (1) whether VIP and galanin were co-expressed by DRG neurons and (2) whether such neurons innervated specified peripheral targets (visceral, cutaneous or muscular). An antibody to the 200 kDa neurofilament subunit (NF200) was used as a marker for large type-A cells in the DRG. VIP and galanin were respectively observed in 22% and 67% of DRG neurons at the L5 spinal level after crushing of the sciatic nerve. Most VIP-containing neurons were small type-B ceils (about 90%) and approximately 95% of VIP-containing neurons also showed galanin-like immunoreactivity. Galanin was expressed by both large type-A and small type-B cells. Immunocytochemistry combined with a retrograde tracer revealed that about 70-80% of the small type-B cells in each sensory division displayed VIP-like immunoreactivity, and that most of the tracer-labeled neurons also expressed galanin. These findings suggest that the expression VIP and/or galanin in response to peripheral nerve crush injury is a property common to visceral, cutaneous and muscular sensory neurons. INTRODUCTION Various neuropeptides have been discovered in rat dorsal root ganglion ( D R G ) cells 11. It is known that the peptide and m R N A levels of vasoactive intestinal peptide (VIP), the VIP-related peptide with amino-terminal histidine and carboxy-terminal isoleucine (PHI, a peptide which is coded on V I P precursor m R N A ) , and galanin increase following nerve crush or transection 8'19'25'26' 30,32. In contrast, these peptides (especially VIP) are present at very low levels in normal D R G neurons. The newly VIP-containing neurons are exclusively small D R G neurons 3°, while galanin-containing neurons are both small and large 8. It is reported that some D R G neurons with injured axons react for both V I P / P H I and galanin antisera 36, but in this case the proportion of coexistence is not known. In the present study, we further investigated the proportions of VIP- and/or galanin-positive neurons and their cell type (large type-A or small type-B) in animals with sciatic nerve crush injuries. It is well documented that D R G neurons contain various combinations of neuropeptides, and that the chemicals expressed are specified by their peripheral targets.
D R G neurons can be divided into three groups acording to their peripheral targets, i.e. those with cutaneous, muscular and visceral target tissues. These peripheral tissues originate from different dermal layers during the developmental process. The chemical expression of D R G neurons which innervate skin and viscera 12'13'22 or skin and muscles 28 have been demonstrated to be different. Also, when their peripheral processes are connected to new locations, peptide expression is altered to become appropriate to the new target 2°. It is thus reasonable to assume that neuronal peptide expression is regulated by target-derived neurotrophic factors (NTFs). Moreover, each neuropeptide seems to be regulated by a specific mechanism involving NTFs. For example, the production of substance P (SP) and somatostatin (SOM) by cultured D R G neurons is regulated by different NTFs 23. Expression of SP and calcitonin gene-related peptide ( C G R P ) in D R G neurons is up-regulated by exogenously administered nerve growth factor (NGF) both in vivo 7 and in vitro 17, while the expression of V I P and galanin seems to be independent of N G F 14'24. We showed in our previous study that blockade of axonal flow at a peripheral site causes an increase in the level
Correspondence: H. Kashiba, Department of Physiology, Kansai College of Acupuncture Medicine, 990 Ogaito, Kumatori, Sennan, Osaka, 590-04, Japan.
48 of these peptides in D R G cells in vivo, suggesting that synthesis of these two peptides in negatively controlled by target-derived NTFs under normal conditions TM. Various kinds of NTFs are tissue specific, such as brain-derived neuro-trophic factor (BDNF) 2 and motor neuron growth factor (MNGF) 2~. Accordingly, the question may be raised as to whether the expression of VIP and galanin is target-related or not. The second aim of this study was to investigate
whether the expression of these peptides in D R G neurons was differentially controlled by their peripheral targets or not. Neurons were labeled by injecting retrograde tracers into the respective peripheral nerve branches innervating skin, muscle and visceral tissues. Subsequently, we examined which neuron groups contained VIP and/or galanin. The distributions of central processes containing newly synthesized V I P or galanin in the dorsal horn was also investigated. A n antibody to the
Fig. 1. Pairs of fluorescent photomicrographs showing VIP-IR (A,C), galanin-IR (B,E) and NF200-IR neurons (D,F) in adjacent serial sections (A and B) and in the same section (C and D, E and F) of the DRG (L5) after crushing of the sciatic nerve. Double-labeled ceils are indicated with arrowheads. Bar = 100 ~m.
49 200 k D a s u b u n i t o f n e u r o f i l a m e n t s ( N F 2 0 0 ) w a s u s e d in t h i s s t u d y as a m a r k e r f o r l a r g e t y p e - A D R G
cells 16'35.
MATERIALS AND METHODS
Tissue preparations Male Wistar rats weighing about 200 g were used. All the animals were divided into three groups and were operated on under sodium pentobarbital anesthesia (50 mg/kg). Those in the first group (n = 4) received crushing of the unilateral sciatic nerve at the hip joint level by tapping it with a pair of forceps for 30 s (about 60 times). The animals in the second group were given injections of a 2% (w/v) solution of fluoro-gold (FG) in distilled water into the unilateral greater splanchnic nerve (0.2/~l) (n = 3), the unilateral sural nerve (the major cutaneous branch of the tibial nerve) (0.4/A) (n = 3), or the unilateral muscular branches of the tibial nerve (0.2 #1 per branch) (n = 3). Injection was performed using a glass micropipette with a tip diameter of 20-50/~m connected to a microinjector. The peripheral regions of the injected nerves were then crushed by tapping with forceps as described above. As a control experiment, animals in the third group (n = 2) received FG injection into the sural nerve or the greater splanchnic nerve and the nerves proximal to the injection site were ligated and cut. All the animals in the three groups were kept alive for 7 days after these treatments. All rats were perfused under deep sodium pentobarbital anesthesia (75 mg/kg) with 0.9% saline (100 ml) followed by cold Zamboni's fixative (400 ml) 37. The spinal cords with the bilateral DRGs corresponding to the spinal levels of the treated nerves (greater splanchnic nerve; T9, sural nerve, muscular branches of the tibial nerve, and sciatic nerve; L5) were excised, post~fixed in the same fixative overnight, and placed in 30% (w/v) sucrose in 0.1 M phosphate buffer for 2 days at 4°C. Spinal cords and DRGs were then cut into sections on a cryostat at -20°C, and collected in small wells or mounted on gelatin-coated glass slides.
(Amersham; 1:250) or TR-conjugated sheep anti-mouse Ig (1:250). The preparations were photographed under a fluorescence microscope with U and G filters so as to visualize only FG and TR fluorescence, respectively. Serial sections of DRGs were cut, and the immunoreactive or FG-labeled cells in every fifth section were counted on enlarged photographic prints. The shortest diameters of immunoreactive neurons labeled by FG were also measured. Neurons in the D R G were divided into four groups according to their size; small (<20/~m in diameter), intermediate (20-30/~m), large (30-45/zm), and giant (>45/~m). In group 3 (control), slidemounted D R G sections were air-dried, coverstipped and examined.
Avidin-biotin-peroxidase complex (ABC) method Immunocytochemical demonstration of VIP in the spinal cord was performed by the method of Hsu et al. 9 using an ABC kit (Vector). After incubation, sections were visualized by reaction with 0.02% (w/v) 3.3'-diaminobenzidine (DAB) and 0.005% (v/v) H20 2 in Tris-HCl buffer for 10-30 min. Then the sections were mounted on gelatin-coated slides, dehydrated, and coverslipped with Permount (Fisher Scientific Co.).
Specificity of the antisera The polyclonal antibodies to galanin and VIP were produced against purified natural porcine galanin and VIP, respectively. The specificity of each of these antibodies was checked by absorption tests. VIP-like immunoreactive (IR) or galanin-IR structures were not stained in sections pre-absorbed with synthetic galanin and VIP at 104 M. The monoclonal antibody to NF200 was raised against 200 kDa porcine NE This antibody has been shown to specifically recognize the 200 kDa subunit of NF in a number of mammalian and avian species 29.
RESULTS
Double-staining o f VIP, galanin and NF200 Double-staining for VIP, galanin and NF200 The indirect immunofinorescent method of Coons 5 was used. For double-staining of VIP and galanin in group 1, adjacent 5-/~m serial sections of the L5 D R G were collected onto separate slides, and were incubated for 2 days at 4°C with rabbit anti-VIP (Ortho Diagnostic Systems; 1:25) and rabbit anti-galanin (Serotec; 1:2,000) antiserum diluted with 0.02 M PBS containing 0.3% Triton X. The sections were then washed in 0.02 M PBS and then incubated overnight at 4°C with fluorescein isothiocyanate (FITC)-conjugated sheep anti-rabbit IgG (Miles; 1:1,000). After a final wash, the samples were air-dried and coverslipped with a glycerine and PBS mixture (1:1). The preparations were viewed and photographed under a fluorescence microscope equipped with a B excitation filter. For double-staining of NF200 and VIP, or NF200 and galanin in group 1, slide-mounted 10-/tm D R G sections were first incubated in a mixture of mouse anti-NF200 monoclonal antibody (ICN Biochemicals, Inc.; 1:10,000) and rabbit anti-VIP (1:25) or rabbit antigalanin (1:2,000) polyclonal antibody.. Samples were then incubated with a mixture of Texas red (TR)-conjugated sheep anti-mouse Ig (Amersham; 1:250) and FITC-conjugated sheep anti-rabbit IgG (1:1,000). Preparations were viewed and photographed under an Olympus fluorescence microscope equipped with G and B excitation filters to assure exclusive visualization of TR and FITC fluorescence, respectively. Preliminary control experiments showed no cross-reactions between the primary and secondary antibodies.
Combined retrograde fluorescent dye tracing and immunofluorescence for V1P~ galanin and NF200 For the immunocytochemical demonstration of VIE galanin or NF200 in group 2, slide-mounted sections (10-/tm thick) or freely floating sections (25-/~m thick) were incubated with anti-VIP, antigalanin or anti-NF200 antiserum as described above. The samples were then incubated with TR-conjugated donkey anti-rabbit Ig
VIP-like immunoreactivity (IR) and galanin-IR were r a r e in t h e c o n t r o l D R G s o n t h e c o n t r a l a t e r a l side ( a b o u t 1 and 3% of the DRG neurons, respectively). These neur o n s w e r e all s m a l l t o i n t e r m e d i a t e i n size. O n t h e o t h e r hand,
many VIP-IR
or galanin-IR neurons
were ob-
s e r v e d in t h e i p s i l a t e r a l D R G s a f t e r n e r v e c r u s h i n g (Figs. 1, 2 a n d 3). T h e n u m b e r
of galanin-IR neurons was
TABLE I
Incidences of VIP-1R, galanin-lR and NF2OO-IR in the visceral, cutaneous and muscular DRG neurons of axotomized rats
VIP
mean_+ S.E.M. Galanin
mean + S.E.M. NF200
mean + S.E.M.
Visceral (n = 3)
Cutaneous (n = 3)
Muscular (n = 3)
58/89 (65%) 55/73 (75%) 29/52 (56%) 6 5 _ 5% 76/82 (93%) 48/54 (89%) 32/35 (91%) 91 + 1% 5/68 ( 7 % ) 6/73 ( 8 % ) 8/69 (12%) 9 + 2%
116/251 (46%) 44/87 (51%) 79/153 (52%) 50 + 2% 104/120 (87%) 61/69 (88%) 66/80 (83%) 86 + 2% 31/109 (28%) 45/139 (32%) 37/101 (37%) 32 + 3%
7/25 (28%) 11/42 (26%) 7/45 (16%) 23 + 4% 21/28 (75%) 31/37 (84%) 27/40 (68%) 76 + 5% 27/37 (73%) 45/71 (63%) 29/37 (78%) 71 + 4%
50
Fig. 2. Pairs of fluorescent photomicrographs showing FG-labeled visceral (A), cutaneous (C) and muscular (E) sensory cells and VIP-IR neurons (B,D,F) in the same sections (A and B, C and D, E and F) of the DRG after nerve crush injury. FG-labeled immunoreactive cells are indicated by arrowheads. Bar = 100/~m.
much greater than that of V I P - I R neurons, with 22 and 67% of the D R G neurons (L5) being positive for VIP and galanin, respectively. The results of double-staining for V I P and galanin are shown in Fig. 1. Ninety-four percent of the V I P - I R neurons (59/63 cells) also displayed galanin-IR and 29% of the galanin-IR neurons (59/202 cells) were positive for VIP. NF200-IR neurons and V I P - I R or galanin-IR neurons in the L5 D R G ipsilateral to the crushed sciatic nerve
are shown in Fig. 1. A b o u t 35% of both the ipsilateral injured and contralateral normal D R G neurons showed NF200-IR. These neurons were intermediate to giant in size. Neurons intensely positive for NF200 were often detected in the ipsilateral D R G s . Some V I P - I R or galanin-IR neurons were also positive for NF200 (Fig. 1), with 9% (12/131 cells) and 43% (43/100 cells) of the NF200-IR neurons positive for V I P and galanin, respectively. Conversely, NF200 was detected in 9% (10/110
51
Fig. 3. Pairs of fluorescent photomicrographs showing FG-labeled visceral (A), cutaneous (C) and muscular (E) sensory cells and galanin-IR cells (B,D,F) in the same sections (A and B, C and D, E and F) of the DRG after nerve crush injury. FG-labeled immunoreactive cells are indicated by arrowheads. Bar = 100/tm.
cells) of V I P - I R neurons and 25% (43/173 cells) of galan i n - I R neurons, respectively.
FG and VIP-IR, galanin-IR, or NF2OO-IR V I P - I R , galanin-IR, and NF200-IR neurons as well as F G - l a b e l e d visceral, cutaneous and muscular afferent neurons are shown in Figs. 2, 3 and 4, respectively. A quantitative evaluation of these neurons is shown in Table I. The distribution of the diameters of the immuno-
reactive neurons labeled with F G fluorescence in Fig. 5, together with the m e a n diameters ___ S.D. Positivity for VIP, galanin and NF200 was relatively constant within each group of neurons, but differed between the neurons innervating different tissues (Table I). Injection of F G into the sural nerve p r o d u c e d fluorescence in a large n u m b e r of D R G cells of various sizes from small to giant, in contrast to injection of F G into the greater splanchnic nerve and the muscular branches
52
Fig. 4. Pairs of fluorescent photomicrographs showing FG-labeled visceral (A), cutaneous (C) and muscular (E) sensory cells and NF200-IR cells (B,D,F) in the same sections (A and B, C and D, E and F) of the DRG after nerve crush injury. FG-labeled immunoreactive cells are indicated by arrowheads. Bar = 100 ~m.
of the tibial nerve (Figs. 2, 3 and 4). Many large to giant muscular sensory neurons showed F G fluorescence, but none of the labeled visceral sensory neurons were giant. In control experiments (group 3), no F G fluorescent cells were observed in the T9 D R G of the rat receiving transection of the greater splanchnic nerve after F G injection or in the L5 D R G of the rat in which the sural nerve was cut after F G injection, showing that F G
was specifically transported via these neuronal processes. V I P - I R was observed in 65 and 50% of the visceral and cutaneous sensory neurons, respectively. Most of these neurons were small to intermediate in size. Twentythree percent of muscular afferent neurons were positive for V I P and these neurons were mainly intermediate to large in size. Neurons of various sizes with F G fluorescence were mostly positive for galanin. G a l a n i n - I R was
301
20
53
visceral 6596 23. - •
l~r~
Galanin visceral 91~ 23.9 _+ 6.5 Hm
~ ~ _ . m
NF200 visceral
9t 31.2 + 6.6 Hm
10
0 • I~ m t--
20
~
lO
. . . .
cutaneous 86g
cutaneous 50%
~
201 10 0
23tmusclar 33.4 + 10.1 Hm , . O
20
40
6"0
muscular 76~
~ 20 Cell
.
40 diameter
60
~
.
.--i
. . . .
cutaneous 32~ 38.9 + 6.6 ~Jm
24.9 + 7.9 lJm
o
r ~ . .
7
_
p
muscular 71~ 42.1 _
m
0
20
+
~
40
60
(ijm)
Fig. 5. Histograms showing the distribution of the cell diameter of VIP-IR (left), galanin-IR (center) and NF200-IR (right) neurons labeled with FG injected into the greater splanchnic nerve (upper), the sural nerve (middle) and the muscular branches of the tibial nerve (lower), respectively. The left axis shows the percentage of immunoreactive cells among the FG-labeled neurons in each sensory division. Arrows on the abscissa indicate the mean diameters in each sensory division.
observed in 91, 86 and 76% of visceral, muscular sensory neurons, respectively. observed in 9, 32 and 71% of visceral, muscular sensory neurons, respectively, rons were all relatively large (Fig. 5).
cutaneous and NF200-IR was cutaneous and and these neu-
VIP-IR or galanin-IR fibers in the dorsal horn Dorsal horn VIP-IR and galanin-IR fibers on the ipsilateral and contralateral sides are shown in Figs. 6 and 7, respectively. Little VIP-IR was observed in the contralateral dorsal horn at the T9 and L5 spinal levels. On the other hand, galanin-IR was intense in laminae I - I I of the dorsal horn at the same levels. After crushing of the greater splanchnic nerve, both VIP-like and galaninlike immunoreactivities were slightly increased in the medial superficial laminae of the T9 dorsal horn (Figs. 6B and 7B). After crushing of the cutaneous (sural) or muscular branches of the tibial nerve, VIP-like and galanin-like immunoreactivities were increased in the central areas of laminae I - I I (Figs. 6D,F and 7D,F). VIP-IR cutaneous fibers were more widely distributed mediolaterally than VIP-IR muscular fibers. A small number of galanin-IR fibers were also observed in the deeper layer (laminae III), where no VIP-IR fibers were found. No apparent changes of both kinds of immunoreactivity were observed in other parts of the dorsal horn. DISCUSSION
The number of galanin-IR neurons appearing after nerve crush was about three times that of VIP-IR neu-
rons. A quarter of the galanin-IR neurons also showed NF200-IR while only about 10% of VIP-IR neurons were positive for NF200, indicating that most VIP-IR neurons may be small type-B cells (which are negative for NF200). Co-expression of VIP and galanin was demonstrated in injured sensory neurons, which confirmed previous observations 36. We further demonstrated that about 95% of VIP-positive neurons also displayed ga!anin-IR, while two-thirds of galanin-IR neurons were negative for VIP. These findings suggest that VIP and galanin may be expressed together by damaged small type-B cells, while only galanin is expressed by some of the larger neurons when their peripheral processes are injured. However, very little is known about the physiological functions of VIP and galanin in injured neurons. An increase of VIP and galanin was demonstrated in visceral, cutaneous and muscular sensory neurons following nerve crush in the present study, but the proportion of VIP-IR neurons differed between these three sensory divisions. The proportions of NF200-IR neurons also differed between each sensory division, as has been previously reported 13'2s. Most muscular sensory neurons contained NF200 but visceral sensory neurons only rarely expressed this substance. The size distribution of these neurons was also different. These results indicate that the proportions of large type-A and small type-B cells were different between each sensory division. Most VIP-IR neurons were small type-B cells, as mentioned above, and the proportion of VIP-positive neurons among the small cells remained constant at 70-80% throughout the three sensory divisions. Galanin-IR neu-
54
~f
Fig. 6. Bright-field photomicrographs showing VIP-IR fibers in the dorsal horn at spinal levels corresponding to the crushed nerves; greater splanchnic nerve (upper), sural nerve (middle), and muscular branches of the tibial nerve (lower). The regions where changes in immunostaining were observed on the ipsilateral side (B,D,F) compared to the contralateral side (A,C,E) are indicated by arrowheads. Bar = 200
ktm. rons also do not appear to specifically innervate one of these types of target tissues, because most D R G neurons in each sensory division expressed galanin (76-91%). In our previous report TM, we suggested that the synthesis of VIP and galanin is controlled by negative feedback from target-derived NTF(s). The present study would suggest that these factors, if they exist, are not tissue-specific. Newly synthesized V I P and galanin may be centrally
and peripherally transported, respectively, because their immunoreactivities or contents increase in both the dorsal horn and in the injured peripheral nerve proximal to the lesion 1'33. Therefore, there have been several hypotheses about their functional roles in the central and peripheral regions. In addition to its central analgesic action 15, VIP has been suggested to stimulate glycogenolysis TM, act as an NTF for glial cells 3 and act as a
55
Fig. 7. Immunofluorescent photomicrographs showing galanin-IR fibers in the dorsal horn at spinal levels corresponding to the crushed nerves; greater splanchnic nerve (upper), sural nerve (middle), and muscular branches of the tibial nerve (lower). The regions where changes in immunostaining were observed on the ipsilateral side (B,D,F) compared to the contralateral side (A,C,E) are indicated by arrowheads. Bar = 200/~m.
vasodilator 34. It may contribute to the process of neuronal regeneration through these diverse functions. On the other hand, galanin has been reported to control the release of V I P 1°, dopamine 27, norepinephrine 31 and acetylcholine 6 in the central nervous system, although none of these transmitters apart from V I P has been identified in D R G neurons. Galanin may regulate the release of VIP from the central branches of the primary sensory
neurons to control the above-mentioned actions of VIP during the process of axonal regeneration. Inhibitory effects of intrathecally administered galanin on VIP-induced facilitation of spinal flexor reflex has been demonstrated in axotomized rats 36. In this context, it is of great interest to show the distribution of central branches containing V I P and galanin in axotomized rats. Sites of projection were quite simi-
56 lar between VIP-IR and galanin-IR fibers, in spite of differences of cell size and type of VIP-IR and galanin-IR neurons in the DRG. It is surprising that these peptide-containing fibers were apparently confined to the laminae I and II. A close association between the functional type of peripheral receptors and area of their afferent terminals within the dorsal horn has been well documented (for review see ref. 4); large afferent fibers originated from type-A neurons connected to low threshold mechanoreceptors in the skin or receptors in intrafusal muscles terminate in the deeper laminae, while A6 and C fibers connected to nociceptors or thermoreceptors are distributed exclusively in laminae I and Iio
where nociceptive-specific dorsal horn neurons are localized. Most of the newly appeared galanin-IR fibers were confined to laminae I and II, although at least a quarter of galanin-IR neurons were type-A in axotomized DRG cells. These findings, taken together, may indicate that DRG neurons which synthesize VIP and/or galanin are mainly involved in thermo-nociceptive transmission mechanism.
REFERENCES
13 Kashiba, H., Senba, E., Ueda, Y. and Tohyama, M., Cell size and cell type analysis of calcitonin gene-relateed peptide-containing cutaneous and splanchnic sensory neurons in the rat, Peptide, 12 (1991) 101-106. 14 Kashiba, H., Senba, E., Kawai, Y., Ueda, Y. and Tohyama, M., Axonal blockade induces the expression of vasoactive intestinal polypeptide and galanin in the rat dorsal root ganglion neurons, Brain Res., in press. 15 Komisaruk, B.R., Banas, C., Mehta, A., Cash, P., Whipple, B. and Jordan, E, Analgesic effect of synthetic fragments of vasoactive intestinal peptide in the rat, Soc. Neurosci. Abst., 16 (1990) 565. 16 Lawson, S.N., Harper, A.A., Harper, E.I., Garson, J.A. and Anderton, B.H., A monoclonal antibody against neurofilament protein specifically labels a subpopulation of rat sensory neurons, J. Comp. Neurol., 228 (1984) 263-272. 17 Lindsay, R.M. and Harmar, A.J., Nerve growth factor regulates expression of neuropeptide genes in adult sensory neurons, Science, 337 (1989) 362-364. 18 Magistretti, P.J., Morrison, J.H., Shoemaker, W.J., Sapin, V. and Bloom, EE., Vasoactive intestinal polypeptide induces glycogenolysis in mouse cortical slices: a possible regulatory mechanism for the local control of energy metabolism, Proc. Natl. Acad. Sci. USA, 78 (1981) 6535-6539. 19 McGregor, G.P., Gibson, S.J., Sabate, I.M., Blank, M.A., Christofides, N.D., Wall, P.D., Polak, L.M. and Bloom, S.R., Effect of peripheral nerve section and nerve crush on spinal cord neuropeptides in the rat: increase VIP and PHI in the dorsal horn, Neuroscience, 13 (1984) 207-216. 20 McMahon, S.B. and Gibson, S., Peptide expression is altered when afferent nerves re-innervate inappropriate tissue, Neurosci. Lett., 73 (1987) 9-15. 21 McManaman, J.L., Crawford, EG., Stewart, S.S. and Appel, S.H., Purification of a skeletal muscle polypeptide which stimulates choline acetyltransferase activity in cultured spinal cord neurons, J. Biol. Chem., 263 (1988) 5890-5897. 22 Molander, C., Ygge, J. and Dalsgaard, C.J., Substance P-, somatostatin- and calcitonin gene-related peptide-like immunoreactivity and fluoride resistant acid phosphatase-activity in relation to retrogradely labeled cutaneous, muscular and visceral primary sensory neurons in the rat, Neurosci. Lett., 74 (1987) 37-42. 23 Mudge, A.W., Effect of chemical environment on levels of substance P and somatostatin in cultured sensory neurons, Nature, 292 (1981) 764-767. 24 Mulderry, P.K. and Lindsay, R.M., Rat dorsal root ganglion neurons in culture express vasoactive intestinal polypeptide (VIP) independently of nerve growth factor, Neurosci. Lett., 108 (1990) 314-320. 25 Nielsch, U. and Keen, P., Reciprocal regulation of tachykininand vasoactive intestinal peptide-gene expression in rat sensory
1 Anand, P., Gibson, S.J., Scaravivlli, F., Blank, M.A., McGregor, G.E, Appenzeller, O., Dhital, K., Polack, J.M. and Bloom, S.R., Studies of vasoactive intestinal polypeptide expression in injured peripheral neurons using capsaicin, sympathectomy and m f mutant rats, Neurosci. Lett., 118 (1990) 60-61. 2 Barde, Y.-A., Edgar, D. and Thoenen, H., Purification of a new neurotrophic factor from mammalian brain, EMBO J., 1 (1982) 549-553. 3 Brenneman, D.E., Neale, E.A., Foster, G.A., d'Autremont, S.E. and Westbrook, G.L., Non-neuronal cells mediate neurotrophic action of vasoactive intestinal peptide, J. Cell. Biol., 104 (1987) 1603-1610. 4 Cervero, E, Dorsal horn neurons and their sensory inputs. In T.L. Yaksh (Ed.), Spinal Afferent Processing, Plenum, New York, (1986) pp. 197-216. 5 Coons, A.H., Fluorescent antibody methods. In J.E Danielli (Ed.), General Cytochemical Method, Academic Press, New York, (1958) pp. 399-422. 6 Fiscone, G., Wu, C.E, Consolo, E, Nordstrom, O., Brynne, N., Bartfai, T., Melander, T. and Hokfelt, T., Galanin inhibits acetylcholine release in the ventral hjppocampus of the rat: histochemical autoradiographic in vivo and in vitro studies, Proc. Natl. Acad. Sci. USA, 88 (1987) 7339-7343. 7 Fitzgerald, M., Wall, ED., Goedert, M. and Emson, EC., Nerve growth factor counteracts the neurophysiological and neurochemical effects of chronic nerve section, Brain Res., 332 (1985) 131-141. 8 Hokfelt, T., Wiesenfeld-Hallin, Z., Villar, M.J. and Melander, T., Increase of galanin-like immunoreactivity in rat dorsal root ganglion cells after peripheral axotomy, Neurosci. Lett., 83 (1987) 217-220. 9 Hsu, S.-M., Raine, L. and Fanger, H., Use of avidin-biotinperoxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabelled antibody (PAP) procedures, J. Histochem. Cytochem., 29 (1981) 577-588. 10 Innoue, T., Kato, Y., Koshiyama, H., Yanaihara, N. and Imura, H., Galanin stimulates the release of vasoactive intestinal polypeptide from perfused hypothalamic fragments in vitro and from periventricular structures into the cerebrospinal fluid in vivo in the rat, Neurosci. Lett., 85 (1988) 95-100. 11 Ju, G., Hokfelt, T., Brodin, E., Fahrenkrug, J., Fischer, J.A., Frey, P., Elde, R.P. and Brown, J.C., Primary sensory neurons of the rat showing calcitonin gene-related peptide immunoreactivity and their relation to substance P-, somatostatin-, galanin-, vasoactive intestinal polypeptide- and cholecystokinin-immunoreactive ganglion cells, Cell Tissue Res., 247 (1987) 417-431. 12 Kashiba, H., Senba, E., Ueda, Y., Tohyama, M., Calbindin D28k-containing splanchnic and cutaneous dorsal root ganglion neurons of the rat, Brain Res., 528 (1990) 311-316.
Acknowledgements. This work was partly supported by a Grantin-Aid from the Ministry of Education, Science and Culture of Japan (03770035).
57
26
27
28
29
30
31
neurons following cut and crush injury, Brain Res., 481 (1989) 25-30. Noguchi, K., Senba, E., Morita, Y., Sato, M. and Tohyama, M., Prepro-VIP and preprotachykinin mRNAs in the rat dorsal root ganglion cells following peripheral axotomy, Mol. Brain Res., 6 (1989) 327-330. Nordstrom, O., Melander, T., Hokfelt, T., Bartfal, T. and Goldstein, M., Evidence for an inhibitory effect of the peptide galanin on dopamine release from the rat median eminence, Neurosci. Leg., 73 (1987) 21-26. O'Brien, C., Woolf, C.J., Fitzgerald, M., Lindsay, R.M. and Molander, C., Difference in the chemical expression of rat primary afferent neurons which innervate skin, muscle or joint, Neuroscience, 32 (1989) 493-502. Shaw, G., Debus, E. and Weber, K., The immunological relatedness of neurofilament proteins of higher vertebrates, Eur. J. Cell Biol., 34 (1984) 130-136. Shehab, S.A.S., Atkinson, M.E. and Payne, J.N., The origin of the sciatic nerve and changes in neuropeptides after axotomy: a double-labelling study using retrograde transport of True blue and vasoactive intestinal polypeptide immunohistochemistry, Brain Res., 376 (1986) 180-185. Tsuda, K., Yokoo, H. and Goldstein, M., Neuropeptide Y and galanin in norepinephrine release in hypothalamic slices, Hy-
pertension, 14 (1989) 81-86. 32 Villar, M.J., Cortes, R., Theodorsson, E., Wiesenfeld-Hallin, Z., Schalling, M., Fahrenkrug, J., Emson, P.C. and Hokfelt, T., Neuropeptide expresion in rat dorsal root ganglion cells and spinal cord after peripheral nerve injury with special reference to galanin, Neuroscience, 33 (1989) 587-604. 33 Villar, M.J., Wiesenfeld-Hallin, Z., Xu, X.-J., Theodorsson, E., Emson, P.C. and Hokfelt, T., Further studies in galanin-, substance P- and CGRP-like immunoreactivities in primary sensory neurons and spinal cord: effects of dorsal rhizotomies and sciatic nerve lesions, Exp. Neurology, 112 (1991) 29-39. 34 Williams, T.J., VIP is more potent than PGE 2 as a vasodilator and oedema potentiator in rabbit skin, Br. J. Pharmacol., 77 (1982) 505-509. 35 Wood, J.N. and Anderson, B.H., Monoclonal antibodies to mammalian neurofilaments, Biosci. Rep., 1 (1981) 263-268. 36 Xu, X.-J., Wiesenfeld-Hallin, Z., Villar, M.J., Fahrenkrug, J. and Hokfelt, T., On the role of galanin, substance P and other neuropeptides in primary sensory neurons of rat: studies on spinal reflex excitability and peripheral axotomy, Eur. J. Neurosci., 2 (1990) 733-743. 37 Zamboni, L. and De Martino, C., Buffered picric acid formaldehyde: a new rapid fixative for electron microscopy, J. Cell Biol., 35 (1967) 148A.