Abnormal development of pacinian corpuscles in double trkB;trkC knockout mice

Neuroscience Letters 410 (2006) 157–161

Abnormal development of pacinian corpuscles in double trkB;trkC knockout mice F. de Carlos a,b , J. Cobo a,b , G. German`a c , I. Silos-Santiago d , R. Laur`a d , J.J. Haro e , I. Fari˜nas f , J.A. Vega e,g,h,∗ a

Departamentos de Cirug´ıa y Especialidades Medico-Quir´urgicas, Universidad de Oviedo, Spain b Istituto Asturiano de Odontolog´ıa, Oviedo, Spain c Dipartimento di MO.BI.FI.PA, Sezione di Anatomia, Universit` a di Messina, Italy d Department of Pharmacology, Vertex Pharmaceuticals Inc., San Diego, CA, USA e Departamento de Morfolog´ıa y Biolog´ıa Celular, Universidad de Oviedo, Spain f Departamento de Biolog´ıa Celular, Universidad de Valencia, Spain g de IUOPA, Universidad de Oviedo, Spain h Departamento de Ciencias M´ edicas, Secci´on de Anatom´ıa y Embriolog´ıa Humana, Facultad de Medicina, Universidad San Pablo – CEU, Madrid, Spain Received 6 May 2006; received in revised form 18 July 2006; accepted 22 July 2006

Abstract Pacinian corpuscles depend on either A␣ or A␤ nerve fibers of the large- and intermediate-sized sensory neurons for the development and maintenance of the structural integrity. These neurons express TrkB and TrkC, two members of the family of signal transducing neurotrophin receptors, and mice lacking TrkB and TrkC lost specific neurons and the sensory corpuscles connected to them. The impact of single or double targeted mutations in trkB and trkC genes in the development of Pacinian corpuscles was investigated in 25-day-old mice using immunohistochemistry and ultrastructural techniques. Single mutations on trkB or trkC genes were without effect on the structure and S100 protein expression, and caused a slight reduction in the number of corpuscles. In mice carrying a double mutation on trkB;trkC genes most of the corpuscles were normal with a reduction of 17% in trkB−/−;trkC+/− mice, and 8% in trkB +/−;trkC −/− mice. Furthermore, a subset of the remaining Pacinian corpuscles (23% in trkB−/−;trkC+/− mice; 3% in trkB+/−;trkC−/− mice) were hypoplasic or atrophic. Present results strongly suggest that the development of a subset of murine Pacinian corpuscles is regulated by the Trk-neurotrophin system, especially TrkB, acting both at neuronal and/or peripheral level. The precise function of each member of this complex in the corpuscular morphogenesis remains to be elucidated, though. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Pacinian corpuscles; Neurotrophins; Trk receptors; Mutant mice

Pacinian corpuscles are complex specialized sensory organs that work as rapidly-adapting low-threshold mechanoreceptors [7] functionally connected with the large myelinated A␣ or A␤ nerve fibers of the large- and intermediate-sized sensory neurons of dorsal root ganglia (DRG; [11]). They consist of a central axon (the extreme tip or dendritic zone of the peripheral axonic process of the pseudo-unipolar sensory neurons) surrounded by differentiated Schwann-related cell forming the inner core, and a well developed outer core and capsule of perineurial origin ∗ Corresponding author at: Departamento de Ciencias M´ edicas, Secci´on de Anatom´ıa y Embriolog´ıa Humana, Facultad de Medicina, Universidad San Pablo – CEU, 28668 BOADILLA DEL MONTE, Madrid, Spain. E-mail address: [email protected] (J.A. Vega).

0304-3940/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2006.07.056

[10,17]. These three main parts of the Pacinian corpuscles are continuous with the axon, Schwann cells and perineurial cells of the nerve fibers and share most of their immunohistochemical characteristics (see [17]). Sensory axons are critical for the development of Pacinian corpuscles, and reciprocal interactions between growing sensory axons and target cells initiate their morphogenesis [12,19]. In the last decade, it has been established that specific subtypes of DRG sensory neurons are dependent upon the neurotrophin/Trk receptor system [6] for development, maintenance and specification of the phenotype. In particular, nerve growth factor (NGF)/TrkA control the nociceptive neurons, neurotrophin-3 (NT-3)/TrkC regulate the proprioceptive neurons, and brainderived neurotrophic factor (BDNF)/TrkB and NT-4/TrkB sup-

158

F. de Carlos et al. / Neuroscience Letters 410 (2006) 157–161

port subpopulations of visceral, cutaneous and joint sensory neurons [3]. Thus, mutations in the genes encoding for these proteins cause selective loss of DRG neuronal fractions and the corresponding sensory formations connected to them at the periphery. Therefore, the animals carrying these mutations are ideal models to analyze the neuronal dependence of the specific types of sensory formations (for references see [4,5,14]). The dependence of Pacinian corpuscles from specific DRG neurons is still elusive, and the available data are contradictories. Initially, they were found to be largely normal in both number and structure in mice with deletions of trkB or trkC genes [2,5]. Nevertheless, a recent study has demonstrated that multiple Trk signals are involved in the development of Pacinian corpuscles, and that they are severely reduced in number in mice deficient for BDNF and NT-3, and TrkA and TrkB [14]. In order to clarify the neuronal dependence of Pacinian corpuscles, we used immunohistochemistry and transmissionelectron microscopy to investigate the impact of single or double-targeted mutations in trkB and trkC genes on the Pacinian corpuscles from the interosseal membrane of mice. TrkB and TrkC positive DRG neurons (see [3]) are the only ones able to supply Pacinian corpuscles with A␣ or A␤ axons [11]. If this hypothesis is true, Pacinian corpuscles should be dramatically reduced in number or should be absent in trkB;trkC mutated animals. The mice included in this study were sampled at the post-natal day 25, because at this time Pacinian corpuscles are fully developed and display their typical structural and immunohistochemical characteristics [1,19]. 25-day-old mice with a single targeted mutation in the trkB or trkC genes were provided by Dr. I. Silos-Santiago, and mice with a double-targeted mutation in trkB;trkC genes were provided by Dr. J. Represa (Universidad de Valladolid, Spain). They were bred out over the C57B1/6 background and genotyped by polymerase chain reaction. Wild-type (n = 5), trkB +/− (n = 4), trkB −/− (n = 6), trkC +/− (n = 5), trkC −/− (n = 6), trkB +/−;trkC +/− (n = 4), trkB −/−;trkC +/− (n = 3); and trkB +/−;trkC −/− (n = 5). Double mutant trkB −/−;trkC−/− mice have a lifespan of 2–3 days [15], a time in which Pacinian corpuscles are not fully developed, and were not included in the study. The animals received an overdose of chloral hydrate and were perfussed transcardially first with a cold solution of 0.9 sodium chloride and then with cold 4% paraformaldehyde in 0.1 M PBS, pH 7.4. The limb segments containing the tarsal joints together with the interosseal membrane were isolated and processed independently. As a rule the left hind-limbs were used for immunohistochemistry, whereas the right ones were used for ultrastructural analysis. The pieces used for immunohistochemistry were processed for routine paraffin embedding, sectioned at 10 ␮m thick at a plane perpendicular to the skin surface, and the sections mounted on gelatin-coated microscope slides. The pieces used for ultrastructural study were processed for resin embedding. The immunohistochemical study was carried out on rehydrated sections rinsed in 0.05 M HCl Tris buffer (pH 7.5) containing 1% bovine serum albumin and 0.1% Triton X-100. Then, the endogenous peroxidase activity (3% H2 O2 ) and non specific binding (25% fetal calf serum) were blocked. Sections were

incubated overnight in a humid chamber at 4 ◦ C with a primary antibody anti-S100 protein (polyclonal raised in rabbit, used diluted 1:1000, Dako, Copenhagen, Denmark; catalogue number Z 0311) diluted in Tris–HCl buffer (0.05 M, pH 7.5) containing 1% bovine serum albumin, 0.2% fetal calf serum and 0.1% Triton X-100. Thereafter, the sections were rinsed in the same buffer, and incubated with peroxidase-labeled sheep antirabbit IgG (Amersham, Buckinghamshire, UK) diluted 1:100, for 1 h at room temperature. Finally, sections were washed and the immunoreaction visualized using 3-3’DAB as a chromogen. For control purposes representative sections were processed in the same way as described above using non-immune rabbit serum instead of the primary antibodies, omitting the primary antibodies in the incubation, or using a specifically preabsorbed antiuserum (DPC, Los Angeles, CA, USA) purchased prediluted (see [1]). For the ultrastructural analysis, the pieces fixed in 4% paraformaldehyde were generously washed in 0.1 M PBS, pH 7.5, post-fixed in 1% osmium tetroxide, and routinely embedded in EPON. Semithin sections (1 ␮m) were obtained, stained with toluidin blue and used for quantitative analysis. The ultra˚ were stained with uranyl acetate and lead thin sections (600 A) citrate and examined and photographed under an electron microscope JEOL-JEM-T8. The Pacinian corpuscles attached to the interosseal membrane were identified following morphological criteria and were quantified in one of every 20 sections (for details see [5]). To avoid counting a Pacinian corpuscle twice, they were counted in sections throughout the terminal segment. In addition to the number of corpuscles, the number of the lamellae of the capsuleouter core, and the mean diameter of the inner-core (including the central axon) were measured. Differences between wild-type and mutated mice were tested using a one-tailed Student’s t-test, and values of P ≤ 0.05 were considered as significant. In all the groups of animals investigated, Pacinian corpuscles were always present, and heterozygous or homozygous single mutations on trkB or trkC genes were without effect on the structure (Fig. 1a and b), immunohistrochemical (Fig. 1e and f) and ultrastructural (data not shown) characteristics of these corpuscles. Also, the number (Table 1) of Pacinian corpuscles slightly differed between wild-type animals and those carrying a single

Table 1 Number of Pacinian corpuscles in the interosseal membrane of 25 days old wildtype mice, and in littermates carrying a single or double mutation in trkB or trkC genes Number Wild-type (n = 5) trkB +/− (n = 4) trkB −/− (n = 6) trkC +/− (n = 5) trkC −/− (n = 6) trkB +/−;trkC +/− (n = 4) trkB −/−;trkC +/− (n = 3) trkB +/−;trkC −/− (n = 5)

18.5 17.6 16.1 18.2 18.1 18.2 15.3 17.1

± ± ± ± ± ± ± ±

Variation (%) 1.7 1.4 1.7* 1.1 0.9 1 1.1* 1.2

0 −5 −13 −2 −2 −2 −17 −8

Hypoplasic (%) 0 0 0 0 0 0 23 3

Numbers in bracket corresponds to the number of animals studied. * P < 0.05 with respect the wild-type littermates.

F. de Carlos et al. / Neuroscience Letters 410 (2006) 157–161

159

Fig. 1. Structure (a–d) and expression of S100 protein immunoreactivity (e–h) in Pacinian corpuscles of 25 days old mice carrying single or double mutations in trkB and trkC genes. Scale bar = 25 ␮m for (a–d); 50 ␮m for (e–h).

mutation on trkB (reduction of 5% and 13% in heterozygous and homozygous, respectively) or trkC (reduction of 2% in both heterozygous and homozygous). At the ultrastructural level, the three main components of the Pacinian corpuscles (the central axon, the inner core, and the outer core-capsule) were well developed, as well as the three main segments (considered form the entry of the nerve fibre to the top ending: preterminal, terminal, and ultraterminal; see [10]) were normally arranged (data not shown). In the single mutated animals, no significant variations were observed in relation to the corresponding wild-type littermates, or among the different mutated groups, with respect to the mean diameter of the inner core (identified because display S100 protein immunoreactiviy) or the number of lamellae forming the outer core and the capsule (Table 2). Most of the Pacinian corpuscles observed in mice carrying a double mutation on trkB;trkC genes showed their typical morphology (Fig. 2a) and immunohistochemical profile (data not shown), although the total number of Pacinian corpuscles was reduced by about 18% in trkB−/−;trkC+/− mice, and 8% in trkB +/−;trkC −/− mice. Minimal variations (around 2%) were detected in trkB+/−;trkC+/− mice (Table 1). Nevertheless, a subset of Pacinian corpuscles (23% in trkB−/−;trkC+/− mice; 3% in trkB+/−;trkC−/− mice) clearly differed from the normality and were hypoplasic or atrophic (Fig. 1c,d,g,h). In fact, they were reduced in size, showed atrophy of the S100 positive immunoreactive structures forming the inner core, and the number of lamellae of the outer core-capsule system were diminished (Table 2). These changes were particularly evident at the ultrastructural level in trkB−/−;trkC+/− animals (Fig. 2b). The

central axon was thicker than in typical corpuscles, and contained less and smaller mitochondria, whereas the axonal spines were inexistent. The inner core lamellae were thicker than in Pacinian corpuscles from wild-type animals, and its number was severely reduced. Typically, at the terminal segment of the Pacinian corpuscles (as in Fig. 2a) the lamellae of the inner core do not completely encircled the axon but rather formed two bilaterally arranged symmetrical halves with a “cleft” dividing

Table 2 Number of lamellae in the outer core-capsule and mean diameter (in ␮m) of the inner core in Pacinian corpuscles of the interosseal membrane of 25 days old wild-type mice, and in littermates carrying a single or double mutation in trkB or trkC genes Number of lamellae of the outer core-capsule (mean ± S.E.) Wild.type (n = 24) TrkB +/− (n = 31) TrkB −/− (n = 22) TrkC +/− (n = 26) TrkC −/− (n = 20) TrkB +/−;TrkC +/− (n = 22) TrkB −/−;TrkC +/− (n = 19) TrkB +/−;TrkC −/− (n = 24)

31.5 29.6 32.3 30.2 31.6 27.4

± ± ± ± ± ±

4.1 5.1 5.2 3.8 6.3 3.9

Diameter of the inner core in (mean ± S.E.) 28.2 29.0 29.1 27.8 28.2 30.6

± ± ± ± ± ±

2.3 1.4 2.3 3.3 1.9 5.9

17.2 ± 3.1*

11.3 ± 3.3*

24.6 ± 5.0*

20.6 ± 2.2*

Numbers in bracket corresponds to the number of Pacinian corpuscles measured. * P < 0.05 with respect the other groups.

160

F. de Carlos et al. / Neuroscience Letters 410 (2006) 157–161

Fig. 2. (a) The terminal segment of most Pacinian corpuscles examined, with independence of the mutation showed a single central axon, surrounded by the concentric hemilamellae of the inner core separated by the cleft; the outer core and the capsule completely encircle these structures. (b) A subset of Pacinian corpucles from trkB−/−;trkC+/− mice were hypoplasic. In these corpuscles, the central axon was thicker and lacked axonal spines; the inner core lamellae were reduced in number, and the cleft was mostly absent. a, central axon; c, capsule; ic, inner core; asterisk, cleft. Original magnification 3.000×.

the inner core into two halves. This arrangement was absent in the hypoplasic-atrophic corpuscles and the “cleft” was mostly absent. Furthermore, in the hypoplasic-atrophic corpuscles the inner core lamellae were not typically packed but delimited large interlaminar spaces containing a reduced number of collagen fibrils and non-fibrilar components of extracellular matrix (Fig. 2b). The outer core and the capsule were basically normal, but the number of lamellae reduced. They consisted of thin cell processes that completely encircle the inner core, and each layer was separated by fibrils and connective tissue (Fig. 2). The present study was designed to investigate and re-examine the effect of the single and double mutations of the trkB and trkC genes in the murine Pacinian corpuscles from the interosseal membrane. Pacinian corpuscles depend on sensory A␣ or A␤ nerve fibres for the development and maintenance of their structural integrity [19]. These fibres are originated from intermediate- and large-sized DRG neurons [11], which are those expressing TrkB and TrkC [3]. We excluded from the study TrkA deficient animals because they show absence of small-sized DRG neurons originating A␦ or C fibres which do not supply Pacinian corpuscles. In the trkB knockout mice, a reduction in the number of Pacinian corpuscles of 5% in trkB+/− mice and 13% in trkB−/− mice has been found, which increased up to 17% in animals with an additional mutation in trkC (trkB−/−;trkC+/−). In a previous study from our laboratory using mice of the same strain and age (25 days) no deficit was observed in trkB−/− mice with respect to trkB+/− and wild-type littermates [5]. Thus, the minimal dis-

crepancies between our two studies might be due to individual variation in the number of corpuscles in the animals studied. In disagreement with our results, Sedy et al. [14] observed a reduction of 48% in the number Pacinian corpuscles of newborn trkB−/− mice, a moment in which the trkB-dependent DRG neurons are still present [15] since neuronal loss in DRG of mice deficient in trkB occurs at the second postnatal week [8,15]. Therefore, if Pacinian corpuscles are directly dependent on TrkB positive DRG neurons, the reduction should be evident in 25 days-old animals but not in newborns. The reason for these discrepancies remains to be investigated. It could be speculated that they are related to and are the result of the genetic manipulation of the mice analyzed, which might affect more specifically to DRG neurons in some cases (animals from our laboratory), and the neuronal and peripheral expression of trkB in some others [14]. In this way TrkB is present in both developing [14] and adult [16] Pacinian corpuscles. Regarding the results in trkC deficient animals, associated or not with trkB mutations, they are on line with previous reports in this topic. Pacinian corpuscles are basically normal in both structure and number ([2,14]; present results), thus, suggesting that only a very few percentage of these structures, if any, is dependent on neurons expressing this receptor. Furthermore, the corpuscular deficit in trkB−/−;trkC+/− deficient animals (17%) is roughly equivalent to the addition of trkB−/− (13%) and trkC+/− (2%) single mutated leading further support to the key role of trkB but not trkC in development of Pacinian corpuscles.

F. de Carlos et al. / Neuroscience Letters 410 (2006) 157–161

A striking finding in our study was that approximately a quarter of the surviving Pacinian corpuscles are hypoplasic or atrophic in trkB−/−;trkC+/− mice, and in a lesser extent in the trkB+/−;trkC−/− ones. The development of Pacinian corpuscles starts prenatally but maturation is not complete until the second or third postnatal week [1]. In this process, there is a critical period absolutely dependent on the axon after which Pacinian corpuscles retain their basic structure, even after long-term denervation [19], although they lack the expression of some proteins, such as S100 protein, depending on the presence of the axon [9]. Therefore, the expression of S100 protein in hypoplasis-atrophic Pacinian corpuscles observed in the trkB;trkC mice strongly suggest that they are innervated but the presence of the axon is not sufficient to fully develop the corpuscle. This claims for a local involvement of the neurotrophin-Trk complex in the morphogenesis of Pacinian corpuscles, as suggest the occurrence of TrkA [18], TrkB [16] and p75NTR [9], but not TrkC [16], in the periaxonic cells of Pacinian corpuscles. Interestingly, in places where mechanoreceptors develop, there is a high expression of neurotrophins mRNAs [13]. Local and not neuronal factors might be also at the basis of the reduction (38%) of Pacinian corpuscles found in newborn TrkA deficient mice [14], since TrkA positive neurons do not support development of Pacinian corpuscles. The impact of the mutation of neurotrophin genes on the Pacinian corpuscles of 14-day-old mice result in a reduction ranging from 43% (NT-3) to 5% (NGF). Also large percentages of corpuscles are lost in double neurotrophin mutant, with an evident cooperation between NT-3, BDNF and NT-4 which suggests a synergistic interaction between NT-3 and one of the neurotrophins that signal through TrkB. Although, the elegant and accurate study by Sedy et al. [14] used animals of different ages the data suggest that the role of NT-3 on Pacinian corpuscles is mediated via TrkB. Taken together, our data strongly suggest that development of a subset of murine Pacinian corpuscles is regulated by the Trk-neurotrophin system, especially TrkB, acting both at neuronal and peripheral, i.e. the target tissues, level. The precise function of each member of this complex in the corpuscular morphogenesis remains to be elucidated, though. Acknowledgements This research was supported by grants from Ministerio de Educacion y Ciencia (SAF2002-03355 and SAF200506235), from Ministerio de Sanidad y Consumo (G03/210 and GO3/167), and From Fundacion la Marato TV3. The authors thank Mr. A. Sanchez Mena for stylistic review of the manuscript.

161

References [1] M. Albuerne, J. De Lavallina, I. Esteban, F.J. Naves, I. Silos-Santiago, J.A. Vega, Development of Meissner-like and Pacinian sensory corpuscles in the mouse demonstrated with specific markers for corpuscular constituents, Anat. Rec. 258 (2000) 235–242. [2] P. Ernfors, K.F. Lee, J. Kucera, R. Jaenisch, Lack of neurotrophin-3 leads to deficiencies in the peripheral nervous system and loss of limb proprioceptive afferents, Cell 77 (1994) 503–512. [3] I. Fari˜nas, Neurotrophin actions during the development of the peripheral nervous system, Microsc. Res. Tech. 45 (1999) 233–242. [4] T. Gonz´alez-Martinez, I. Fari˜nas, M.E. del Valle, J. Feito, G. Germana, J. Cobo, J.A. Vega, BDNF, but not NT-4, is necessary for normal development of Meissner Pacinian corpsucles, Neurosci. Lett. 377 (2005) 12–15. [5] T. Gonzalez-Martinez, G.P. Germana, D.F. Monjil, I. Silos-Santiago, F. de Carlos, G. Germana, J. Cobo, J.A. Vega, Absence of Meissner corpuscles in the digital pads of mice lacking functional TrkB, Brain Res. 1002 (2004) 120–128. [6] E.J. Huang, L.F. Reichart, Trk receptors: roles in neuronal signal transduction, Ann. Rev. Biochem. 72 (2003) 609–642. [7] K.O. Johnson, The roles and functions of cutaneous mechanoreceptors, Curr. Opin. Neurobiol. 11 (2001) 455–461. [8] R. Klein, R.J. Smeyne, W. Wurst, L.K. Long, B.A. Auerbach, A.L. Joyner, M. Barbacid, Targeted disruption of the trkB neurotrophin receptor gene results in nervous system lesions and neonatal death, Cell 75 (1993) 113–122. [9] S.M. Lopez, M. Perez-Perez, J.M. Marquez, F.J. Naves, J. Represa, J.A. Vega, p75 and TrkA neurotrophin receptors in human skin after spinal cord and peripheral nerve injury, with special reference to sensory corpuscles, Anat. Rec. 251 (1998) 371–383. [10] L. Malinovsky, Sensory nerve formations in the skin and their classification, Microsc. Res. Tech. 34 (1996) 283–301. [11] E.R. Perl, Function of dorsal root ganglion neurons: an overview, in: S.A. Scott (Ed.), Sensory Neurons: Diversity, Development and Plasticity, Oxford University Press, New York, 1992, pp. 3–23. [12] R. Saxod, Ontogeny of the cutaneous sensory organs, Micros. Res. Tech. 34 (1996) 313–333. [13] L.C. Scehcterson, M. Bothwell, Novel roles for neurotrophins are suggested by BDNF and NT-3 mRNA expression in developing neurons, Neuron 9 (1992) 449–463. [14] J. Sedy, V. Szeder, J.M. Walro, Z.G. Ren, O. Nanka, M. Sieber-Blum, M. Grim, J. Kucera, Pacinian corpuscles development involves multiple Trk signalling pathways, Dev. Dyn. 231 (2004) 551–563. [15] I. Silos-Santiago, A.M. Fagan, M. Garber, B. Fritzsch, M. Barbacid, Severe sensory deficits but normal CNS development in newborn mice lacking TrkB and TrkC tyrosine kinase receptors, Eur. J. Neurosci. 9 (1997) 2045–2056. [16] B. Stark, M. Risling, T. Carlstedt, Distribution of the neurotrophin receptors p75 and TrkB in pheripheral mechanoreceptors; observations on changes after injury, Exp. Brain Res. 136 (2001) 101–107. [17] J.A. Vega, J.J. Haro, M.E. Del Valle, Immunohistochemistry of human cutaneous Meissner and Pacinian corpuscles, Micros. Res. Tech. 34 (1996) 351–361. [18] J.A. Vega, E. Vazquez, F.J. Naves, B. Calzada, M.E. Del Valle, J.J. Represa, Immunohistochemical localization of the high-affinity NGF receptor (p140-trkA) in adult dorsal root and sympathetic ganglia, and in the nerves and sensory corpuscles supplying digital skin, Anat. Rec. 240 (1994) 579–588. [19] J. Zelena, Nerves and Mechanoreceptors, Chapman & Hall, London.