distribution and chemical coding of sensory neurons innervating … · 2016. 12. 20. ·...

Neuropeptides xxx (2016) xxx–xxx

YNPEP-01757; No of Pages 14

Contents lists available at ScienceDirect

Neuropeptides

j ourna l homepage: www.e lsev ie r .com/ locate /npep

Distribution and chemical coding of sensory neurons innervating the skin of theporcine hindlimb

Anna Kozłowska a,⁎, Anita Mikołajczyk b, Zbigniew Adamiak c, Mariusz Majewski a

a Department of Human Physiology, Faculty of Medical Sciences, University of Warmia and Mazury in Olsztyn, Polandb Department of Public Health, Epidemiology and Microbiology, Faculty of Medical Sciences, University of Warmia and Mazury in Olsztyn, Polandc Department of Surgery and Radiology, Faculty of Veterinary Medicine, University of Warmia and Mazury in Olsztyn, Poland

⁎ Corresponding author at: Department of Human PSciences, University of Warmia and Mazury in Olsztyn, WPoland.

E-mail address: [email protected] (A. Kozł

http://dx.doi.org/10.1016/j.npep.2016.10.0040143-4179/© 2016 The Authors. Published by Elsevier Ltd

Please cite this article as: Kozłowska, A., et alNeuropeptides (2016), http://dx.doi.org/10.1

a b s t r a c t
a r t i c l e i n f o
Article history:Received 1 March 2016Received in revised form 17 October 2016Accepted 24 October 2016Available online xxxx

The aim of the present study was to establish the origin and chemical phenotyping of neurons involved in skininnervation of the porcine hind leg. The dorsal root ganglia (DRGs) of the lumbar (L4–L6) and sacral (S1–S3) spinalnerves were visualized using the fluorescent tracer Fast Blue (FB). The morphometric analysis of FB-positive(FB+)neurons showed that in the L4, L5, S1 and S2 DRGs, the small-sized perikarya constituted the major popu-lation, whereas in the L6 and S3 DRGs themedium-sized cells made up the major population. In all these ganglia,large-sized FB+ perikarya constituted only a small percentage of all FB+ neurons. Immunohistochemistry re-vealed that small- andmedium-sized FB+ perikarya contained sensory markers such as: substance P (SP), calci-tonin gene related peptide (CGRP) and galanin (GAL); as well as various other factors such as somatostatin(SOM), calbindin-D28k (CB), pituitary adenylate cyclase-activating polypeptide (PACAP) and neuronal nitricoxide synthase (nNOS). Meanwhile large-sized FB+ perikarya usually expressed SP, CGRP or PACAP. Inthe lumbar DRGs, some large cells also contained SOM and CB. Double-labeling immunohistochemistryshowed that SP-positive neurons co-expressed CGRP, GAL or PACAP; while PACAP-positive cells co-expressed GAL or nNOS. Neurons stained for SOM were also immunoreactive for CB or GAL, while neuronsstained for nNOS were also immunoreactive for GAL. In conclusion, the present data has indicated that thedistribution and chemical phenotyping of the porcine skin-projecting neurons are different within DRGs ofthe lumbar (forming a femoral nerve) and sacral (forming a sciatic nerve) spinal nerves.

© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords:SkinDRGChemical codingGilts

1. Introduction

Pseudounipolar afferent neurons localized in dorsal root ganglia(DRGs) have two branches extending from the common trunk: a pe-ripheral process innervating the somatic or visceral target and the cen-tral branch conveying information to the spinal cord or brainstem(Gilbert, 2000; Rowe and Iwamura, 2001). In humans, as well as inother species that have been studied so far (rat, mouse and cat), skin-projecting neurons supplying the lower leg/hind paw were distributedin a distinct set of DRGs. For example, in humans such neurons werefound in L1–S3 DRGs, in rodents they were found within the L4–L6DRGs while in the cat they were localized within the L6–L7 and S1DRGs (Barabas and Stucky, 2013; Mclachlan and Jänig, 1983; Minett etal., 2014; Royden et al., 2013; Takahashi et al., 2003).

From a physiological point of view, skin-projecting DRGs neuronsmay be divided into fourmain subtypes: (i) nociceptors, sensing painful

hysiology, Faculty of Medicalarszawska 30, 10-082 Olsztyn,

owska).

. This is an open access article under

., Distribution and chemical co016/j.npep.2016.10.004

(nociceptive) stimuli, (ii) pruriceptors, conveying itch sensations, (iii)thermoceptors, registering temperature information and (iv) a huge,quite heterogeneous group of low-threshold mechanoreceptors(LTMRs), encoding touch – i.e., awide variety of non-painfulmechanicalstimuli (McGlone and Reilly, 2010). However, each of these physiologi-cally-defined neuronal subsets could be further separated into differentfunctional subtypes on the basis of various histological as well as phys-iological parameters such as their soma size, axon diameter, degree ofmyelination as well as conduction velocity (for a review see Abrairaand Ginty, 2013). For example, pig DRG cells were divided into small(b30 µm) to medium-sized (b50 μm) and large perikarya (N50 μm;Bossowska et al., 2009) according to the diameter of their soma, whiletheir axons were classified as A and C fibers (Harper and Lawson,1985). However, it should be stressed that each of the populations ofneuronal processes should, in fact, be considered as a set of terminalsemerging from functionally distinct heterogeneous groups of DRG neu-rons. For example, the population of “Aβ fibers” (an anatomical sub-strate for LTMRs) could be further divided into at least four functionalsubtypes, distinguished primarily by their ability to adapt to sustainedmechanical stimulation (rapidly-adapting vs. slowly-adapting – RA vs.SA, respectively). These four functional subtypes would be: (i) Merkel

the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

ding of sensory neurons innervating the skin of the porcine hindlimb,

0opyright_ulicense

http://dx.doi.org/10.1016/j.npep.2016.10.004

mailto:[email protected]

Journal logo


0opyright_ulicense

http://www.sciencedirect.com/science/journal/

www.elsevier.com/locate/npep


2 A. Kozłowska et al. / Neuropeptides xxx (2016) xxx–xxx

cells-innervating Aβ SA1-LTMRs, responsible for the recognition of thestatic nature of the touch stimuli, (ii) Ruffini's corpuscles-related AβSA2-LTMRs, sensing the stretching of skin, (iii) Meissner's corpuscles-supplying Aβ RA1-LTMR fibers, sensitive to movements across theskin and (iv) Aβ RA2-LTMRs, terminating in Pacinian corpuscles andsensing high-frequency vibrations (Zimmerman et al., 2014). However,these subtypes do not convey nociceptive stimuli (Johnson, 2008). Aδterminals locatedwithin the skin aremainly responsible for sensingme-chanical (especially pinprick or pinching), thermal and chemical stimu-li; these afferents have also been shown to be involved in hyperalgeticstates in mice (Byers and Bonica, 2001; Yeomans et al., 2004). Small-sized neurons that form the source of intradermal C fibers constitutethe major population of skin-supplying DRG cells (Smith and Lewin,2009). Since these neurons respond to thermal, chemical (acute andchronic itch) as well as nociceptive stimuli, they were therefore termedpolymodal cells (Johnson, 2008). Furthermore, although it is now gen-erally accepted that virtually all touch sensations are mediated by theabove-described fast conducting Aβ LTMRs, there is a growing body ofevidence that a population of C fibers (termed “C-tactile afferents”)play a crucial role in conveying the so-called “pleasure touch stimuli”generated during grooming or nurturing behaviors (McGlone andReilly, 2010; Zimmerman et al., 2014). Thus, according to the above-mentioned data, it is widely accepted that the vast majority of small-sized (and at least part of the medium-sized) sensory neurons partici-pate in nociception/pruriception as well as thermal stimuli sensing(Djouhri et al., 2003; Petruska et al., 2000). Meanwhile, large-sizedperikarya are mainly involved in mechanoreceptive and/or propriocep-tive information transmission (Le Pichon and Chesler, 2014; Li et al.,2016).

Immunohistochemical studies have demonstrated that the small-sized skin-projecting DRG neurons may be roughly divided into twogroups. The first group is the “non-peptidergic” population, whichbinds Bandeiraea simplicifolia lectin (I-B4) and utilizes glutamate as itsmain transmitter (Plenderleith and Snow, 1993). This group is mostlikely involved in itch processing (for a review see Ma, 2010). The sec-ond group is the “peptidergic” population, which exhibits numerousbioactive molecules that are widely accepted as markers of nociceptivecells (Ye et al., 2013). These substances include, among others: (i) sub-stance P (SP), a peptide responsible not only for neurogenic dermatitis,but also for wound healing (for a review see Blais et al., 2014), (ii) calci-tonin gene-related peptide (CGRP; Taddese et al., 1995), a peptide thatapart from its involvement in sensory transmission, appears to act as avery prominent regulatory agent of the immunological and inflammato-ry reactions of the skin, interactingwith the antigen-presenting capabil-ity of Langerhans cells (Granstein et al., 2015), (iii) somatostatin (SOM;Molander et al., 1987) which is most likely expressed by a subset ofpruriceptors (Stantcheva et al., 2016) as well as nociceptors (Shi et al.,2014), (iv) galanin (GAL; Kashiba et al., 1994) responsible not only fornociceptive (mainly thermal) transmission, but also for the regenerativeabilities of sensory cells, especially the peripheral, skin-projecting ter-minals of them (Hill et al., 2010), (v) nitric oxide synthase (NOS;Wang et al., 2013), a marker for NO synthesizing afferent cells, contrib-uting to mechanical hypersensitivity during the maintenance phase ofneuropathic pain (Kim et al., 2011), (vi) pituitary adenylate cyclase-ac-tivating polypeptide (PACAP; Moller et al., 1993), involved not only inpain processing, but also in the neuroprotection of challengedDRG neu-rons (McIlvain et al., 2006) or (vii) calbindin-D28k (CB; Duc et al.,1994), a calcium-binding protein being a specific marker to a distinctpopulation of capsaicin-resistant skin-projecting neurons (Kashibaet al., 1990).

It must be stressed that there is still a lack of data concerning boththe distribution pattern aswell as the neurochemical features of porcineskin-projecting DRGs neurons supplying the hindlimb. Such data couldbe of great importance because the pig, as opposed to rodents (rats,mouse and guinea pig) or carnivores (cat), shows distinct anatomical,physiological, biochemical and immunological similarities to humans

Please cite this article as: Kozłowska, A., et al., Distribution and chemical coNeuropeptides (2016), http://dx.doi.org/10.1016/j.npep.2016.10.004

(Avon and Wood, 2005). Therefore, the pig should be considered to bean especially valuable species for studying human skin diseases, suchas dermal melanoma, hypertrophic scarring, bullous pemphigoid andallergic and contact dermatitis among others (Herron, 2009; Hrubanet al., 2004; Olivry et al., 2000; Vana and Meingassner, 2000; Zhu etal., 2003). Furthermore, although the organization of the dermal inner-vation in the pig is still not fully elucidated, porcine skin has also beenused as a model to study mechanism(s) of cutaneous pain, wound andburn lesion healing as well as radiation and laser exposure (DiGiminiani et al., 2013; Kim et al., 2013; Levi et al., 2011; Sullivan et al.,2001). Therefore, the present study was aimed at (i) revealing thesources of origin of afferent fibers involved in skin innervation of theporcine hindlimb by the use of retrograde tract-tracing, (ii) defining ofthe chemical phenotypes of traced neurons by means of single- anddouble-immunofluorescence labeling and (iii) comparing the inter-and intraganglionic distribution pattern, as well as the neurochemicalfeatures of retrogradely-labeled neurons that have sent their processesto cutaneous branches of the femoral and sciatic nerves.

2. Materials and methods

2.1. Animals

Four juvenile female crossbred gilts (Pietrain x Duroc), aged 8–12 weeks and weighing 15–20 kg were used. All animals were housedand treated in accordance with the Principles of Laboratory AnimalCare (NIH publication no. 86-23, revised 1985). All experimental proce-dures were approved by the Local Ethics Commission of the Universityof Warmia and Mazury in Olsztyn (no. 70/2012). During the entire ex-periment, all efforts were made to minimize both the number of ani-mals used as well as their suffering. Seven days before the beginningof the experiment, the animals were transported from a farm to thelocal animal facility where they were individually housed in stallsunder conditions of natural light and room temperature. All animalswere fed with a commercial grain mixture and tap water ad libitum.24 h before surgery feeding was stopped.

2.2. Anesthesia, surgery and tissue processing

Thirty minutes before the main anesthetic (sodium thiobarbital;Thiopenthal, Sandoz, Austria; 20 mg/kg b.w., i.v., in a fractionated infu-sion) was given, the animals were pretreated with atropine sulfate(Polfa, Poland; 0.04 mg/kg b.w., s.c.) and azaperone (Stressnil, JanssenPharmaceutica, Belgium; 2.0mg/kg b.w., i.m). After induction of the sur-gical anesthesia, the skin of the left hindlimb was gently shaved,disinfected with a 1% water-alcohol solution of iodine tincture andthen a chessboard-style grid was drawn on it with a dermatologicallyneutral, permanentmarker; the sides of the resulting squaresmeasuredapproximately 1 cm. By the use of a Hamilton syringe equipped with a26 gauge needle, 1 μl of 5% aqueous solution of fluorescing retrogradeneuronal tracer Fast Blue (FB; EMS-Chemie GmbH, Groß-Umstadt,Germany) was injected intradermally into the center of each square.For example, 50 injections was made in the area innervated by thecutenous branches of the femoral nerve (50 μl) and for the sciaticnerve another 50 (50 μl). This procedure allowed for the complete cov-erage of the skin regions innervated by cutaneous branches of the fem-oral and sciatic nerves, respectively (Fig. 1a). Four weeks later allanimals were euthanized by an overdose of sodium thiobarbital and,after the cessation of breathing and heart beat, transcardially perfusedwith an initial flush of a preperfusion solution containing 0.9% NaCl(POCH, Poland), 2.5% polyvinylpyrrolidone (Sigma, Germany), 0.05%procaine hydrochloride (Polfa, Poland) and 20,000 IU of heparine(Polfa, Poland; added just before the perfusion started), followed by4% paraformaldehyde (Merck, Germany) in 0.1 M phosphate buffer(pH 7.4). Th12–S4 DRGs ipsilateral to the tracer injection sites were col-lected from all studied animals and then postfixed by immersion in the



Fig. 1. a. A schematic diagram of the intradermal injection of Fast Blue in areas of porcine skin, hindlimb innervated by cutaneous branches of the femoral and sciatic nerves. Dark purplerepresents the area of the cutaneous branches of the femoral nerve (lateral femorocutaneousnerve) and light purple represents the cutaneous branches of the sciatic nerve (including tibaland peroneal nerve; Sobociński, 1973). b. A schematic diagram of dorsal root ganglia (DRG) showing topographical subdomains in which the relative frequency of small-, medium- andlarge-sized skin projecting neurons were counted; cr-c– cranio-central, cr-p– cranio-peripheral, ca-c – caudo-central and ca-p – caudo-peripheral subdomains of the DRG. (Forinterpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3A. Kozłowska et al. / Neuropeptides xxx (2016) xxx–xxx

same fixative for 15 min, washed twice in 0.1 M phosphate buffer(pH=7.4, 4 °C) over three days and then stored in 18%buffered sucrosesolution containing 0.01% sodiumazide (pH=7.4) at 4 °C until freezingand sectioning.

2.3. Immunofluorescence procedures

Prior to sectioning, each DRGs studied was mounted on a small padformed from frozen Tissue-Tek OCT compound (Sakura Finetek, USA) ina way fully resembling its anatomical orientation, allowing the propermaintenance of the cranio-caudal, dorso-ventral as well as centro-pe-ripheral axis of the studied ganglion. Ten-μm-thick serial cryostat(HM525 Zeiss, Germany) sections were mounted on chrome alum-gelatin coated glass slides and, in order to determine thepresence of ret-rogradely traced cells in particular DRGs, sections selected from threedifferent levels of each ganglion (dorsal, middle and ventral one-third,respectively), were shortly examined under the epi-fluorescence-equipped microscope (see below). This allowed for the establishmentof the range of DRGs involved in porcine hind leg skin innervation.Based upon the results of this procedure, routine single- or double-im-munofluorescence labeling was applied to each fourth section of L4–L6and S1–S3 DRGs. It should be stressed that these ganglia were used forfurther study owing to the fact that their containedmost of retrogradelylabeled neurons.

All steps of the staining procedureswere performed in a humid cham-ber protected against UV light (in order to prevent the photobleaching ofFB-labeled neurons). In brief, frozen sections were always air-dried atroom temperature (RT) for 45min, rinsed three times in phosphate-buff-ered saline (PBS, pH 7.4), and then incubated for 1 h with a blocking


solution containing 1% Triton X-100, 0.1% bovine serum albumin, 0.05%thimerosal (all fromSigma-Aldrich, St. Louis,MO), 0.01%NaN3 (POCH, Po-land) and 10% normal goat serum (Jackson Immunoresearch, USA) in0.01 M PBS (pH 7.4). This solution was then used as a diluent of primaryas well as secondary reagents listed in Table 1.

For the single-labeling immunofluorescence, sections were rinsedagain in PBS and incubated overnight (RT) with a solution of primaryantibody and then, after repetitive rinse in PBS, the reaction was visual-ized by incubation (RT; 1 h) with an appropriate secondary antiserumconjugated to either fluorescein isothiocyanate (FITC), Alexa or biotin.In the latter case, the reaction was visualized by a subsequent incuba-tionwith CY3-conjugated streptavidin (RT; 1 h). For the double-labelingimmunofluorescence, a mixture of two primary antibodies raised indifferent host-species were used. After the incubation with primary an-tibodies, sections were rinsed in PBS (3 × 15min) and incubated for 1 hwith a mixture of appropriate FITC- or Alexa-conjugated secondary an-tiserum and biotinylated donkey anti-rabbit antibody. The latter anti-body was finally visualized by additional incubation of sections withstreptavidin-CY3 complex for 1 h. Following subsequent rinsing in PBS(3 × 15min), sections were coverslippedwith carbonate-buffered glyc-erol (pH 8.6).

An additional set of labelings was done in order to disclose both thepresence as well as relative frequency of afferent skin-projecting neu-rons that were immunonegative to all of the antisera in the study. Toreach this goal, a mixture of all primary antibodies used in this experi-ment (with an emphasis on the preservation of their respectiveworkingdilutions)was applied to the sections for a routine overnight incubation,and, after a thoroughwash in PBS, visualized using the appropriatemix-ture of secondary antibodies labeled with CY3.



Table 1Immunoreagents used.

Primary antisera

AntigenAntibodycode Host species Dilution Supplier

SP 8450-0505 Rat 1:400 AbDSerotecSP 8450-0004 Rabbit 1:3000 AbDSerotecCGRP AB43873 Rabbit 1:18,000 AbcamGAL T-5036 Guinea pig 1:6000 BachemGAL AB5909 Rabbit 1:16,400 Merck

MilliporenNOS N2280 Mouse 1:600 Sigma

AldrichnNOS AB5380 Rabbit 1:13,000 Merck

MilliporePACAP H-052-02 Rabbit 1:11,000 PhoenixSOM MAB354 Rat 1:300 Merck

MilliporeCB AB1778 Rabbit 1:11,000 Merck

MilliporeNeuN MAB377 Mouse 1:1000 Merck

Millipore

Secondary reagentsReagent Conjugated

toAntibodycode

Dilution Supplier

Donkey anti-guinea pigIgG (H + L)

FITC 706-095-148 1:1000 Jackson

Donkey anti-mouse IgG(H + L)

FITC 715-095-151 1:1000 Jackson

Donkey anti-rat IgG (H + L) FITC 712-095153 1:1000 JacksonGoat anti-Mouse IgG (H + L)Secondary Antibody, AlexaFluor® 488 conjugate

Alexa A-11001 1:1000 ThermoFisherScientific

Biotinylated goat anti-rabbitimmunoglobulins

biotin E0432 1:1500 Dako

Streptavidin CY3 016-160-084 1:10,000 Jackson

Abbreviations: SP - substance P, CGRP - calcitonin gene related peptide, GAL - galanin,nNOS - neuronal nitric oxide synthase, PACAP - pituitary adenylate cyclase-activatingpolypeptide, SOM – somatostatin, CB - calbindin-D28k, NeuN - neuronal nuclei andFITC - fluorescein isothiocyanate.


2.4. Specificity tests of tracing and labeling procedures

Traces of FB were neither found inmuscles, nor in the subcutaneoustissues in the vicinity of the tracer deposition locations. The specificity ofprimary and secondary antisera was tested by: (i) preabsorption testsbased on the incubation of sections with antibody that had beenpreabsorbed with synthetic antigen (25 μg of appropriate antigen per1 ml of corresponding antibody at working dilution) and (ii) omissionand replacement tests during which the primary or secondary antibodywas omitted or replaced by non-immune sera or PBS, respectively. Lackof any visible fluorescence indicated specificity of the labeling.

2.5. Counting of neurons and statistical analysis

Labeled sections were viewed under an Olympus BX61 microscopeequipped with epi-fluorescence and appropriate, small-band filter setsfor FB (V1 module; excitation filter 330–385 nm, barrier filter420 nm), FITC (B1 module; excitation filter 450–480 nm, barrier filter515 nm) and CY3 (G1module; excitation filter 510–550 nm, barrier fil-ter 590 nm). Microphotographs were acquired using a 20× UPlanSApoobjective (0.75 NA) and a PC equipped with a CCD camera operated bycellSens Dimension image analyzing software (Olympus, Poland). Allphotomicrographs presented in this work have been published “as is”without any alterations being done to the image quality (contrast,brightness) using image editing programs. In order to avoid double-counting of labeled neurons, only FB-positive (FB+) perikarya with aclearly visible nucleuswere counted in each fourth section of the partic-ular DRG studied. The perikarya were divided into three subclasses:


small (up to 30 μm); medium (31–50 μm) and large (N 50 μm) accord-ing to Bossowska et al. (2009). Their mean size was measured usingcellSens Dimension image analyzing software.

To verify the intraganglionic distribution of FB+ neurons, each ofthe DRGs sections of interest was divided along its long and short axisinto four subdomains: cranio-central, cranio-peripheral, caudo-centraland caudo-peripheral (Fig. 1b), and each of the counted FB+ cells wasadditionally assigned to one of the above-mentioned subdomains.Small-, medium- and large-sized FB+ neurons were counted withineach subdomain of each individual lumbar and sacral DRGs studied.These counts were expressed as the percentage (relative frequencies)of small-, medium- and large-sized FB+ neurons in each of the studiedganglia and/or subdomains, always considering the total number ofFB+ neurons as being 100% within the studied DRG or subdomain.A similar paradigm was used to evaluate the inter- as well asintraganglionic distribution patterns of the single- and double-immunolabeled FB+ neurons. However, the relative frequencies of neu-rons exhibiting a particular antigen (or combination of antigens) werecalculated from the pooled values of each class size of FB-labeled neu-rons in all lumbar (L4–L6) or sacral (S1–S3) DRGs, respectively. Finally,raw counts from each animal were pooled and presented as amean ± standard error of mean (SEM) values.

The statistical importance of differences found between groups ofdata were analyzed using one-way analysis of variance (ANOVA)followed by the Tukey test using GraphPad Prism 4 software (GraphPadSoftware, La Jolla, CA, USA). Data on the number of skin-projectingperikarya in the total population of lumbar and sacral DRGs wereanalyzed using the Mann-Whitney U test. P b 0.05 was considered tobe statistically significant.

3. Results

3.1. The distribution pattern of sensory skin-projecting neurons within theporcine dorsal root ganglia (DRGs)

As revealed by the retrograde tract-tracing procedure, the porcineneurons supplying skin of the hindlimb were distributed in dorsal rootganglia (DRGs) from the lumbar (L; L4-L6) to the sacral (S; S1-S3) spinalnerves. It is worth noticing that these Fast Blue-positive (FB+) neuronswere much more numerous within the sacral DRGs than their counter-parts in lumbar DRGs (Table 2). For example, the mean number of FB+neurons from L4 to S3 DRGs was 1324 ± 61 and 63.2 ± 0.4% of theseneuronswere located in the sacral DRGs. Especially dense accumulationof the FB+ neurons was found in S2 where they constituted 37.5± 0.9%of the total population of DRGs cells. It should be added that a few FB+

neuronswere also found in other DRGs:mainly in L2, L3 and S4 – unpub-lished data.

Three subpopulations of DRG neurons have been observed: small-(the mean diameter 19.7 ± 0.9 μm), medium- (the mean diameter42.4 ± 0.5 μm) and large- (the mean diameter 65.0 ± 1.4 μm) sized.The proportions among those subpopulations varied considerably de-pending upon the location of the DRGs according to the vertebral levelas well as the distribution of these neurons inside the ganglia.

The rostro-caudal distribution of small-, medium- and large-sizedFB+ neuronswas very similar within the lumbar and sacral DRGs. Actu-ally, the pattern of distribution in L4-L6 DRGs was subsequently repeat-ed in S1-S3 DRGs (Fig. 2). For example, in L4 and S1 DRGs the smallperikarya constituted the most numerous populations of FB+ neuronswhile the medium- and large-sized neurons were much less numerous(L4 - 63.7 ± 1.7%, 33.4 ± 1.9%, 2.8 ± 0.4% and S1-63.1 ± 2.5%, 32.1 ±2.3%, 4.7 ± 0.3%; respectively). In L5 DRG the percentages of the small-sized neurons were significantly higher compared to medium- andlarge-sized cells (L5 - 51.6±2.2%, 43.1±1.5%, 5.2±0.4%; respectively).Meanwhile, in S2 DRG the percentages of small- andmedium-sized neu-rons were similar, but significantly higher than the percentages of thelarge neurons (S2 - 48.3 ± 1.1%, 45.2 ± 0.9%, 6.4 ± 0.1%; respectively).



Table 2The percentages contribution of skin-projecting perikarya in the total population of lum-bar (L; L4–L6) and sacral (S; S1–S3) dorsal root ganglia (DRGs). Data are presented as themean (range) of 4 animals.

Marker

DRG

L4–L6 S1–S3

FB+NeuN+/FB-NeuN+ 3.8 ± 0.1 (96/2598) 5.4 ± 0.4 (120/2498)⁎

The absolute number of cells containing markers is in brackets.Abbreviations: FB - Fast Blue; NeuN - neuronal nuclei.⁎ Indicates differences (P b 0.008) between the lumbar and sacral DRGs.

a

b

c

Small perikarya

Medium perikarya

Large perikarya

0

10

20

30

40

50

*

cr-p cr-c cd-ccd-p

*

Fre

qu

en

cy

o

f F

B-la

be

le

d c

ells

[%

]

0

10

20

30

40

50

cr-p cr-c cd-ccd-p

aa

aa

L

S

Fre

qu

en

cy

o

f F

B-la

be

le

d c

ells

[%

]

10

20

30

40

50

a

a

a

(P<0.01)

a

***

******

***

eq

ue

nc

y o

f F

B-la

be

le

d c

ells

[%

]


In L6 and S3 DRGs, the percentages of the medium-sized neurons werethe highest when compared to the percentage value of small- as wellas large-sized neurons (L6 - 50.9 ± 0.8%, 32.9 ± 1.8%, 16.0 ± 1.4% andS3 - 50.2 ± 2.5%, 33.8 ± 0.8%, 15.9 ± 0.9%; respectively). It is worthmentioning that in L6 and S3 DRGs the percentages of the large-sizedcells were also significantly lower than medium ones.

When the distributions of the small-, medium- and large-sized FB+neurons were compared within 4 subdomains of lumbar and sacralDRGs, the results were as follows (Figs. 3a-c).

The distribution patterns of the small-sized neurons were similar inlumbar and sacral DRGs and percentages of these cells within thesubdomains of ganglia did not differ statistically (Fig. 3a). Furthermore,those cells were always the most numerous in the cranio-peripheral(cr-p; L4–L6 - 39.5 ± 1.0%; S1–S3 - 37.5 ± 2.3%) and caudo-peripheral(cd-p; L4–L6 - 32.7 ± 1.0%; S1–S3 - 35.3 ± 1.2%) subdomains. On theother hand, the lowest percentages were noted in the cranio-central(cr-c; L4–L6 - 13.1 ± 0.8%; S1–S3 - 13.6 ± 0.5%) and caudo-central (cd-c; L4–L6 - 14.5 ± 0.4%; S1–S3 - 13.5 ± 2.2%) subdomains.

The percentage of themedium-sized neurons (Fig. 3b)was higher inthe cd-c subdomains of S1–S3 DRGs (33.9 ± 1.7%) as compared to L4–L6DRGs (15.5 ± 0.6%). Meanwhile, the percentage of cells in the cr-psubdomain was higher in the L4–L6 DRGs (31.6 ± 2.5%) compared tothe S1–S3 DRGs (19.3 ± 3.2%). The comparison of cr-c (L4–L6 - 23.3 ±2.5% vs S1–S3 - 19.6 ± 3.4%) and cd-p (L4–L6 - 29.4 ± 1.1% vs S1–S3 -27.1 ± 6.1%) subdomains of L4–L6 and S1–S3 DRGs has shown that thepercentages do not differ significantly.

The large neurons have not shown a single pattern of distributionwithin the lumbar and sacral DRGs and the percentages were signifi-cantly different in each studied segment (Fig. 3c). For example, in thelumbar DRGs, these cells were significantly more numerous in cr-c(L4–L6 33.5 ± 2.6% vs S1–S3 15.3 ± 0.3%) and cd-p subdomains (L4–L6

0

25

50

75

100

Nu

mb

er o

f F

B-lab

ele

d ce

lls [%

]

small

medium

large

L4 L

5 L

6 S

1 S

2 S

3

ganglia

a

***

a

***

***

***

***

a

***

aa

a

***

***

***

***

***

Fig. 2. A bar diagram showing the segmental distribution of skin-projecting sensoryneurons following injection of Fast Blue (FB) into the skin. Bars represent meanpercentages of the neurons located in the left ganglia. ***– indicates differences(P b 0.001) between small- and medium- as well as large-sized perikarya; a – indicatesdifferences (P b 0.001) between the medium- and large-sized perikarya. Error barsrepresent ±SEM; n = 4.

0

cr-p cr-c cd-ccd-p

Fr

Fig. 3. A bar diagram showing the distribution of skin-projecting sensory neurons inparticular subdomains of the DRG following injection of Fast Blue (FB) into the skin ofthe pig. Bars represent mean percentages of the neurons located in the lumbar (L; L4–L6) and sacral (S; S1–S3) DRGs. *P b 0.05; ***P b 0.001 - indicates differences between thelumbar and sacral DRGs; a - indicates differences (P b 0.01; P b 0.001) between thecranio-peripheral (cr-p) and cranio-central (cr-c) or caudo-peripheral (cd-p) andcaudo-central (cd-c) subdomains of the lumbar and sacral DRGs. Error bars represent±SEM; n = 4.


37.3 ± 2.7% vs S1–S3 13.8 ± 2.0%) when compared to sacral DRGs.Meanwhile, sacral DRG cells were significantly more numerous in cr-p(S1-S3 30.8 ± 1.2% vs L4–L6 13.3 ± 1.1%) and cd-c (S1–S3 40.3 ± 3.8%vs L4–L6 15.5 ± 1.1%) subdomains when compared to lumbar DRGs.

3.2. Immunohistochemistry of porcine dorsal root ganglia (DRG)

All details concerning neurochemical phenotypes of FB+ sensoryskin-projecting neurons within the porcine DRGs were demonstratedin Tables 3 and 4. The morphology and characteristic features of single-and double-immunolabeled structures were visualized in Figs. 4-7.




3.3. Single labeling immunohistochemistry

Single labeling immunohistochemistry revealed that the majority ofsmall-, medium- and large-sized FB+ displayed immunoreactivity tosubstance P (SP), calcitonin gene-related peptide (CGRP) and pituitaryadenylate cyclase activating polypeptide (PACAP). However, the per-centages of small-, medium- and large-sized cells expressing these sub-stances were significantly different (Table 3). Small-sized FB+ neurons(Figs. 4a, c, d, 6a and e)were especially rich in SP (Figs. 4a′, c′, d′, 6a′ ande′), CGRP (Figs. 4a″ and 6a″) and PACAP (Fig. 4d″). N80% of these cellsexpressed SP and N90% of them expressed CGRP. The percentage ofPACAP containing cells was around 70% (Figs. 5c″ and 6e″- devoid ofPACAP).Medium-sized FB+neurons (Figs. 4e and 6b)were significantlyless often immunoreactive to SP (Figs. 4e′ and 6b′), CGRP (Fig. 6b″) andPACAP (Fig. 4e″), than small-sized cells. For example, only around 40%of these neurons expressed SP and around 60% of them expressedCGRP. The percentage of PACAP containing medium-sized cells wasaround 55%. Large FB+ neurons (Figs. 4b and 6d) were the least immu-noreactive to SP (devoid of SP: Figs. 4b′ and 6d′), CGRP (devoid of CGRP:Fig. 4b″) and PACAP, when compared to the small- and medium-sizedcells. b20% of neurons expressed SP and only around 35% of themexpressed CGRP or PACAP.

It is interesting to note that there were significant differences be-tween the L4–L6 and S1–S3 DRGs as far as the percentages of galanin(GAL), somatostatin (SOM), neuronal nitric oxide synthase (nNOS), orcalbindin (CB) expressing FB+ cells were concerned. For example,GAL (Figs. 4c″, 5a″, c′, e′, 6c″, 7a″, c′ and e′ - devoid of GAL) was foundin N21% of small FB+ neurons (Figs. 4c, 5a, c, e, 6c, 7a, c and e), in 13%of medium-sized cells and only single large-sized neurons (Fig. 6d)were immunoreactive to GAL (Fig. 6d″ – devoid of GAL).

In the case of SOM and CB, the differences between the percentagesof small- (FB+/SOM+: Figs. 5a/a′, b/b′, 7a/a′ and b/b′; FB+/CB+: Figs. 5b/b″ and 7b/b″), and medium-sized FB+ neurons has shown a similar

Table 3The percentages of different populations among FB-labeled skin-projecting small-; medium- aganglia (DRGs). Data are presented as the mean (range) of 4 animals.

Studied substances amongFB-labeled neurons

Small perikarya Medium

L S L

SP+/FB+ 92.5 ± 1.5 79.4 ± 4.8 37.2 ± 4.6a,(Pb0.001SP−/FB+ 7.5 ± 0.9 20.6 ± 6.3 62.8 ± 5.6a,(Pb0.001TS+/FB+ (94) 190/202 (81) 89/110 (39) 31/79CGRP+/FB+ 93.7 ± 2.6 90.1 ± 2.4 72.7 ± 4.7a,(Pb0.05)CGRP−/FB+ 6.3 ± 4.0 9.9 ± 2.9 27.3 ± 7.7TS+/FB+ (96) 199/208 (93) 102/110 (76) 72/95PACAP+/FB+ 72.8 ± 5.8 67.9 ± 6.0 53.5 ± 7.8PACAP−/FB+ 27.2 ± 7.8 32.1 ± 8.9 46.5 ± 11.3TS+/FB+ (76) 112/147 (71) 80/112 (56) 78/139GAL+/FB+ 21.1 ± 1.3 26.2 ± 0.6 13.1 ± 0.9a,(Pb0.001GAL−/FB+ 78.9 ± 6.9 73.8 ± 4.6 86.9 ± 2.6TS+/FB+ (23) 39/167 (29) 28/97 (14) 27/189SOM+/FB+ 33.2 ± 1.3 13.3 ± 1.1⁎,(Pb0.001) 17.8 ± 13.8SOM−/FB+ 66.8 ± 7.2 86.7 ± 6.7 82.2 ± 2.9TS+/FB+ (35) 55/156 (20) 32/159 (19) 22/113nNOS+/FB+ 37.9 ± 2.3 44.4 ± 7.6 0.9 ± 0.5a,(Pb0.01)

nNOS−/FB+ 62.1 ± 9.8 55.6 ± 12.9 99.1 ± 0.8a,(Pb0.05)

TS+/FB+ (40) 45/113 (48) 52/107 (4) 4/104CB+/FB+ 15.6 ± 0.9 4.6 ± 0.5⁎,(Pb0.001) 3.7 ± 1.2a,(Pb0.001)CB−/FB+ 84.4 ± 3.9 95.4 ± 2.1 96.3 ± 1.8TS+/FB+ (18) 17/96 (9) 8/88 (5) 6/122

VR - was very rarely observed.TS+/FB+ − the absolute number of cells containing tested substances (TS+; percent in brackeThe total number of cells counted was 3541.Abbreviations: SP - substance P, CGRP - calcitonin gene relatedpeptide,GAL – galanin, SOM – somand nNOS – neuronal nitric oxide synthase.⁎ Indicates differences (P b 0.05; P b 0.001) between the lumbar and sacral DRGs for the sama Indicates differences (P b 0.05; P b 0.01; P b 0.001) between the small- and medium-sizedb Indicates differences (P b 0.05; P b 0.01; P b 0.001) between the small- and large-sized pec Indicates differences (P b 0.05) between the medium- and large-sized perikarya within th


distribution pattern, with only some large FB+ neurons expressingSOM and CB.

About 40% of small- (Figs. 5e/e″, 7d/d′ and e/e″) and medium-sizedFB+ neurons (Figs. 5d/d′ and 7d/d′ excluding the medium-sized cellsfrom lumbar DRGs) contained nNOS, whereas the large-sized FB+ neu-ronswere devoid of thismarker. It is worth noticing that therewere sig-nificant differences between the lumbar and sacral DRGs in thepercentages of SOM and CB (small-sized neurons) and nNOS (medi-um-sized neurons) expressing FB+ cells.

3.4. Double labeling immunohistochemistry

Double labeling immunohistochemistry gave additional insight intothe neurochemical phenotypes of small-, medium- and large-sized FB+

neurons (Table 4).Almost all small-sized FB+ neurons (Figs. 4a, c, d, 6a and c) express-

ing SP (Figs. 4a′, d′, 6a′ and c′)were simultaneously CGRP- (Figs. 4a″ and6a″) or PACAP-positive (Fig. 4d″) and N20% of them co-expressed GAL(Figs. 4c″ and 6c″). Among the medium-sized FB+/SP+ neurons(Figs. 6b/b′), the level of co-localization with CGRP (Fig. 6b″) and GALwas similar to that found in small-sized neurons. Also, the percentagesof FB+/SP+/PACAP+medium-sized neurons (Figs. 4e/4e′/4e″) were sig-nificantly lower when contrasted to small-sized neurons (Figs. 4d/4d′/4d″) while those containing GAL were more numerous in sacral DRGsthan in lumbar DRGs. Large-sized FB+/SP+ neurons co-expressedCGRP mainly in sacral DRGs. They were sporadically immunoreactiveto GAL and completely devoid of PACAP. The percentages of small-sized FB+/SP+ neurons co-expressing CGRP, PACAP or GALwere similarin lumbar and sacral DRGs. On the other hand, the percentages of medi-um-sized cells with the same phenotypes were statistically different inthose DRGs.

Small- (Figs. 5a, b, 7a and b) and medium-sized FB+ neurons con-taining SOM (Figs. 5a′, b′, 7a′ and b′) also co-expressed GAL (Figs. 5a″

nd large-sized perikarya located in the lumbar (L; L4–L6) and sacral (S; S1–S3) dorsal root

perikarya Large perikarya

DRG

S L S

) 43.5 ± 6.0a,(Pb0.01) 9.0 ± 3.3b,(Pb0.001); c,(Pb0.05) 23.8 ± 9.9b,(Pb0.001)

) 56.5 ± 9.6a,(Pb0.05) 91.0 ± 4.8b,(Pb0.001) 76.2 ± 10.1b,(Pb0.01)(46) 44/95 (11) 3/28 (31) 6/19

59.6 ± 5.5a,(Pb0.001) 43.7 ± 7.8b,(Pb0.01) 31.8 ± 13.6b,(Pb0.001)40.4 ± 8.5a,(Pb0.01) 56.3 ± 11.4b,(Pb0.05) 68.2 ± 18.1b,(Pb0.01)

(62) 54/87 (45) 5/11 (33) 6/1857.1 ± 4.7 37.1 ± 18.8 30.2 ± 12.742.9 ± 6.8 62.9 ± 5.7 69.8 ± 13.3(60) 63/105 (48) 12/25 (45) 10/22

) 13.2 ± 0.9a,(Pb0.001) VR VR86.8 ± 2.7 0 0(15) 20/131 VR VR

6.6 ± 1.0a,(Pb0.01) 4.7 ± 2.8 VR93.4 ± 2.3 95.3 ± 4.7 VR(8) 10/128 (5) 1/18 VR

44.4 ± 7.6⁎,(Pb0.001) 0 055.6 ± 2.3⁎,(Pb0.05) 0 0

(47) 45/95 0 03.6 ± 2.1 6.1 ± 3.5 096.4 ± 3.6 93.9 ± 6.1 0(6) 7/124 (9) 2/22 0

ts) and FB+ cells (FB+).

atostatin, CB - calbindin-D28k, PACAP - pituitary adenylate cyclase-activating polypeptide

e population of small- or medium-sized perikarya.perikarya within lumbar and sacral DRGs.rikarya within lumbar and sacral DRGs.e lumbar and sacral DRGs.



Table 4Percentage of Fast Blue (FB) retrogradely labeled skin-projecting small-;medium- and large-sized perikarya located in the lumbar (L; L4–L6) and sacral (S; S1–S3) ganglia of the dorsal rootganglia (DRGs) coexpressing the studied substances in different combinations. Data are presented as the mean (range) of 4 animals.

Studied substances amongFB-labeled neurons

Small perikarya Medium perikarya Large perikarya

DRG

L S L S L S

SP+/CGRP+ 97.3 ± 1.3 99.3 ± 0.6 97.0 ± 1.8 80.9 ± 2.7⁎,(P b 0.05); a,(P b 0.01) VR 52.3 ± 3.9b,(P b 0.001); c,(P b 0.001)SP+/CGRP− 1.0 ± 0.3 0.3 ± 0.1⁎,(P b 0.05) 1.2 ± 0.9 5.1 ± 0.8a,(Pb0.01) VR 6.6 ± 2.6SP−/CGRP+ 1.1 ± 0.2 0.4 ± 0.1 1.3 ± 0.5 1.5 ± 0.7 VR 16.7 ± 9.2b,(Pb0.01); c,(Pb0.01)SP−/CGRP− 0.6 ± 0.1 0 0.5 ± 0.2 12.5 ± 1.3⁎,(P b 0.01) VR 24.4 ± 2.1c,(P b 0.05)TS+/FB+ (99) 106/107 (100) 145/145 (99) 159/161 (89) 88/98 VR (79) 23/29SP+/GAL+ 21.5 ± 2.2 26.9 ± 2.7 6.6 ± 0.2a,(P b 0.05) 34.2 ± 1.7⁎,(P b 0.001) VR VRSP+/GAL− 62.9 ± 3.7 58.8 ± 2.3 37.0 ± 5.7a,(P b 0.01) 21.8 ± 5.6a,(P b 0.001) VR VRSP−/GAL+ 2.8 ± 1.4 0 2.4 ± 2.0 4.6 ± 0.8*,(P b 0.05) VR VRSP−/GAL− 12.8 ± 5.2 14.3 ± 4.9 54.0 ± 5.7a,(P b 0.001) 39.4 ± 5.5a,(P b 0.01) VR VRTS+/FB+ (89)140/157 (88) 107/122 (48) 84/175 (63) 106/168 VR VRSP+/PACAP+ 93.4 ± 0.8 91.7 ± 3.5 43.9 ± 1.0a,(P b 0.001) 56.6 ± 0.9⁎,(P b 0.05); a,(P b 0.001) 0 0SP+/PACAP− 2.2 ± 0.4 1.5 ± 0.2 31.4 ± 3.5a,(P b 0.001) 15.7 ± 3.5⁎,(P b 0.01);a,(P b 0.05) VR VRSP−/PACAP+ 3.2 ± 0.8 2.8 ± 0.4 4.6 ± 0.4 4.5 ± 0.5 VR VRSP−/PACAP− 1.2 ± 0.6 4.0 ± 0.8 20.1 ± 5.0a,(P b 0.05) 23.2 ± 6.1a,(P b 0.05) VR VRTS+/FB+ (99) 96/97 (97) 71/73 (83) 73/88 (79) 126/160 VR VRSOM+/GAL+ 39.8 ± 3.5 47.2 ± 4.3 45.7 ± 4.4 33.1 ± 0.7 VR VRSOM+/GAL− 14.1 ± 3.5 8.9 ± 4.4 9.9 ± 3.6 8.6 ± 4.9 VR VRSOM−/GAL+ 9.5 ± 4.9 13.7 ± 4.9⁎,(P b 0.01) 3.7 ± 1.1a,(P b 0.05) 0.8 ± 0.8a,(P b 0.05) VR VRSOM−/GAL− 36.6 ± 6.5 30.2 ± 4.9 40.7 ± 8.8 57.5 ± 2.0a,(P b 0.05) VR VRTS+/FB+ (66) 67/101 (72) 69/96 (63) 79/125 (47) 50/106 VR VRSOM+/CB+ 25.1 ± 2.5 26.5 ± 3.1 33.8 ± 1.5a,(Pb0.01) 32.8 ± 2.9a,(P b 0.05) 0 VRSOM+/CB− 20.4 ± 3.7 13.1 ± 7.3 8.3 ± 3.6 5.5 ± 3.1 VR VRSOM−/CB+ 5.3 ± 2.4 3.0 ± 3.0 0 1.7 ± 1.7 VR VRSOM−/CB− 49.2 ± 6.8 57.4 ± 7.4 57.9 ± 2.2 60.0 ± 2.1 VR VRTS+/FB+ (46) 43/93 (55) 67/122 (46) 70/152 (41) 64/157 VR VRPACAP+/GAL+ 74.8 ± 3.2 69.3 ± 2.2 3.0 ± 1.7a,(P b 0.01) 15.0 ± 1.4⁎,(P b 0.05); a,(P b 0.01) 0 VRPACAP+/GAL− 12.9 ± 5.1 17.5 ± 6.6 58.7 ± 4.8a,(P b 0.01) 53.5 ± 8.3a,(P b 0.01) VR VRPACAP−/GAL+ 8.2 ± 3.0 8.5 ± 2.5 0 1.2 ± 0.8 VR VRPACAP−/GAL− 4.1 ± 1.5 4.7 ± 2.1 38.3 ± 4.0a,(P b 0.01) 30.3 ± 6.2a,(P b 0.05) VR VRTS+/FB+ (99) 74/75 (99) 85/86 (64) 70/109 (72) 61/84 VR VRPACAP+/nNOS+ 48.9 ± 1.4 62.6 ± 9.5 50.2 ± 3.1 34.7 ± 2.5⁎,(P b 0.01); a,(P b 0.05) 0 0PACAP+/nNOS− 39.0 ± 0.9 26.2 ± 4.6 32.6 ± 6.8 33.7 ± 6.0 VR VRPACAP−/nNOS+ 5.9 ± 3.0 2.5 ± 1.3 3.1 ± 2.1 6.9 ± 3.3 0 0PACAP−/nNOS− 6.2 ± 1.3 8.7 ± 3.4 14.1 ± 3.4a,(Pb0.05) 24.7 ± 4.2a,(Pb0.05) VR VRTS+/FB+ (96) 126/131 (93) 100/107 (78) 117/149 (80) 133/165 VR VRnNOS+/GAL+ 8.2 ± 0.20 0 0 28.3 ± 1.8 VR 0nNOS+/GAL− 36.4 ± 4.2 29.6 ± 3.9 0.8 ± 0.8a,(P b 0.01) 1.6 ± 0.8a,(P b 0.05) 0 0nNOS−/GAL+ 25.0 ± 6.4 28.5 ± 4.8 31.5 ± 6.1 39.2 ± 5.3 VR VRnNOS−/GAL− 30.4 ± 4.4 41.9 ± 6.1 67.7 ± 5.9a,(P b 0.01) 30.9 ± 5.1⁎,(P b 0.05) VR VRTS+/FB+ (72) 65/90 (61) 37/60 (34) 65/189 (72) 62/86 VR VR

VR – was very rarely observed.TS+/FB+ - the absolute number of cells containing tested substances (TS+; percent in brackets) and FB+ cells (FB+).The total number of cells counted was 3844.Abbreviations: SP - substance P, CGRP - calcitonin gene relatedpeptide,GAL – galanin, SOM – somatostatin, CB - calbindin-D28k, PACAP - pituitary adenylate cyclase-activating polypeptideand nNOS – neuronal nitric oxide synthase.⁎ Indicates differences (P b 0.05; P b 0.01; P b 0.001) between the lumbar and sacral DRGs for the same populations of small-, medium- or large-sized perikarya.a Indicates differences (P b 0.05; P b 0.01; P b 0.001) between the small- and medium-sized perikarya within the lumbar or sacral DRGs.b Indicates differences (P b 0.001) between the small- and large-sized perikarya within the lumbar or sacral DRGs.c Indicates differences (P b 0.05; P b 0.01; P b 0.001) between the medium- and large-sized perikarya within the lumbar or sacral DRGs.


and 7a″) or CB (Figs. 5b″ and 7b″) while the large-sized FB+/SOM+

cells were very rarely immunoreactive to those substances (Table4). The pattern of distribution of small- and medium-sized FB+/SOM+/GAL+ neurons was nearly the same in lumbar and sacralDRGs. In the case of FB+/SOM+/CB+ neurons, the percentage ofsmall–sized perikarya was significantly lower than medium-sizedcells.

Small-sized FB+ neurons (Fig. 5c) containing PACAP (Fig. 5c″) co-expressed GAL (Fig. 5c′) much more often than medium-sized cells. Insacral DRGs the percentage of medium-sized FB+/PACAP+/GAL+neurons was significantly higher than in lumbar DRGs. The extent ofco-expression of PACAP andNOS in small- andmedium-sized FB+neu-rons (Figs. 7d/d′/d″) was quite similar (excluding the medium-sizedcells from lumbar DRGs). Small-sized FB+/PACAP+/GAL+ andFB+/PACAP+/NOS+ neurons were homogenously distributed


within lumbar and sacral DRGs, in contrast to their medium-sizedcounterparts. Large FB+/PACAP+ cells were almost always devoidof GAL or nNOS.

Among FB+ neurons expressing nNOS there was always a sub-population of small- (Figs. 5e/e′/e″), medium- and large-sized cellswhich were simultaneously immunoreactive to GAL. Interestingly,such a subpopulations were represented in the lumbar DRGs bysmall-sized neurons, while in the sacral DRGs by medium-sizedneurons.

4. Discussion

This is the first report that provides a detailed description ofboth the distribution pattern, as well as the chemical phenotypes ofporcine skin-projecting afferent neurons supplying the hindlimb.



Fig. 4.Micrographs showing immunohistochemical staining of the Fast Blue (FB)-labeled neurons in the DRGs L4–L6. The small-sized arrows show positive immunoreactivity, while large-sized arrows show thenegative immunoreactivity of the skin-projecting neurons of the examined substances. All of the imageswere taken separately from blue (a, b, c, d, e), green (a′, b′, c′,d′, e′) and red (a″, b″, c″, d″, e″) fluorescent channels. a. One FB+ small-sized neuronwhichwas simultaneously a′ SP+ and a″ CGRP+. b. Two FB+ small- and large-sized neuronswhichwere negative for b′ SP and b″ CGRP. c. Three small-sized FB+neuronswhichwere simultaneously c′ SP+ and c″GAL+, and one FB+medium-sized neuronwhichwas simultaneously c′SP+ and c″ GAL−. d. A small group of small-sized FB+ neurons which were simultaneously d′ SP+ and d″ PACAP+. e Three FB+medium-sized neurons which were simultaneously e′SP+ or SP− and e″ PACAP+. Scale bar = 50 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)


4.1. The inter- and intraganglionic distribution pattern of skin-innervatingafferent neurons

The results of the present study showed that in the pig the skin-projecting afferent neurons supplying hind leg were distributed inL4–S3 DRGs, what roughly corresponds with data obtained in human(L1–S3; Royden et al., 2013), rodents (rat and mouse; L4–L6 DRGs;Barabas and Stucky, 2013, Minett et al., 2014, Takahasi et al., 2003)and cat (L6–L7 and S1; McLachlan and Jänig, 1983). However, in rodentssuch neurons were exclusively present in lumbar DRGs, while inhuman, cat and, especially, in the pig, they were also present in sacralganglia, constituting in the latter speciesmore than half (approximately63%; present study) of all retrogradely-labeled neurons.

We have also observed that in the pig the skin-projecting DRG neu-rons may be subdivided into the three “size-classes”, with the smallneurons forming the most numerous populations within L4 and S1DRGs, andmedium cells predominating in L6 and S3 DRGs. The percent-age of large-sized neurons was always significantly lower when com-pared to small- and medium-sized cells. These results are in partcomparable to the findings reported by Bossowska et al. (2009) and


Russo et al. (2013), although, these authors studied the population ofthe primary sensory DRG neurons projecting to the urinary bladder infemale and male pigs, respectively.

There are other interesting observations as far as the distributionpattern of the porcine skin-innervating DRG neurons is concerned. Forexample, it appears that in the pig the distribution pattern of small-and medium-sized neurons within lumbar DRGs is subsequently re-peated within the sacral DRGs, a phenomenon not described in anyother species studied so far. However, as may be judged from resultsof Wessells et al. (1990), such “repetitive” distribution could reflectthe origin of a skin-oriented afferent component of the femoral and sci-atic nerves, a suggestion supported byfindings that small- andmedium-sized neurons were mainly responsible for transmission of nociceptiveand thermal stimuli (for a review see Abraira and Ginty, 2013;Petruska et al., 2000).Meanwhile, large-sized neuronsmay be predictedto be mechanoreceptors and/or proprioceptors which transfer stimula-tion via thickly myelinated afferent Aβ-fibers (Li et al., 2016). This as-sumption needs to be verified experimentally, especially since there isa lack of any data addressing this topic in available literature. Anotherinteresting feature of the distribution pattern of the porcine skin-



Fig. 5.Micrographs showing immunohistochemical staining of the Fast Blue (FB)-labeled neurons in the DRGs L4–L6. The small-sized arrows show positive, while large-sized arrows shownegative immunoreactivity of the skin-projecting neurons in theDRG to examined substances. All of the imageswere taken separately from blue (a, b, c, d, e), green (a′, b′, c′, d′, e′) and red(a″, b″, c″, d″, e″) fluorescent channels. a. One FB+ small-sized neuron which was simultaneously a′ SOM+ and a″ GAL+. b. One FB+ small-sized neuron immunoreactive for b′ SOM+and b″ CB+ andmedium-sized neuron devoid of both studied substances (b′ SOM− and b″ CB−). c. One FB+ small-sized neuronwhichwas simultaneously c′GAL+ and c″ PACAP+. d.One FB+medium-sized neuronwhichwas simultaneously d′NOS+ and d″ PACAP+. e. Two FB+ small-sized neurons: onewas simultaneously e′GAL+ and e″NOS+while the secondwas devoid of both studied substances (e′ GAL- and e″ NOS-). Scale bar =50 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the webversion of this article.)


innervating DRG neurons was their somatotopic arrangement withinsubdomains of the studied ganglia, determined by ganglion geometry(i.e., the arrangement of its longest and shortest axis). It has beenfound that small neurons were localized mainly in the cranio- andcaudo-peripheral subdomains of studied DRGs,while, the topographicalarrangement of medium and large neurons was more complicated.For example, the majority of medium neurons was found in cranio-pe-ripheral and caudo-central subdomains (lumbar and sacral DRGs, re-spectively). Large neurons were primarily found in caudo-peripheral(lumbar DRGs) and caudo-central subdomains (sacral DRGs). As men-tioned above, a cranio-caudal somatotopic organization of skin-projecting DRG neurons supplying the rat hindlimb has previouslybeen postulated (Wessells et al., 1990), with the suggestion that lumbarDRG neurons located within cranial ganglionic domains projected intothe femoral nerve, while neurons observed in the caudal domains con-tributed to the sciatic nerve. Although the cranio-caudal arrangement ofskin-projecting porcine afferent cells observed in the present studymaybe indicative for a similar organization of the sensory components of


above-mentioned nerve trunks in the pig, this suggestion needs to beverified in detail.

4.2. Chemical phenotyping of skin-innervating afferent neurons and itsfunctional implications

As revealed in the present study, the majority of skin-projectingneurons supplying porcine hind leg, regardless of their diameter,expressed either SP, CGRP or PACAP, while the subpopulations of GAL-, SOM-, CB- or NOS-immunoreactive (-IR) cells were distinctly less nu-merous. When both the soma diameter and the chemical phenotypingof these cells were taken into consideration, it has been found that SP-, CGRP- or PACAP-containing cells predominated within the subset ofsmall- and medium-sized neurons, while among large neurons suchphenotypes were rare. Similar observations concerning SP- and CGRP-IR skin-projecting neurons were observed in the rat (Aoki et al., 2005),mouse (Suzuki et al., 2010) and guinea pig (Gibbins et al., 1985). Itshould be pointed out that there is a lack of data concerning PACAP-IR



Fig. 6.Micrographs showing immunohistochemical staining of the Fast Blue (FB)-labeled neurons in the DRG S1-S3·The small-sized arrows show positive, while large-sized arrows shownegative immunoreactivity of the skin-projecting neurons in theDRG to examined substances. All of the imageswere taken separately from blue (a, b, c, d, e), green (a′, b′, c′, d′, e′) and red(a″, b″, c″, d″, e″)fluorescent channels. a. Two small-sized FB+neuronswhichwere simultaneously a′ SP+and a″CGRP+. b. Onemedium-sized FB+neuronwhichwas simultaneously b′SP+ and b″ CGRP+. c. Two FB+ small-sized neurons: onewas simultaneously c′ SP+ and c″GAL+while the secondwas devoid of both studied substances (c′ SP– and c″GAL−). d. Onelarge-sized FB+neuronwhichwas devoid of d′ SP and d″GAL. e. One small-sized FB+neuronwhichwas e′ SP+ and e″ PACAP−. Scale bar=50 μm. (For interpretation of the referencesto colour in this figure legend, the reader is referred to the web version of this article.)


cells. The available reports focused only on the general presence of suchcoded perikarya in dorsal root and trigeminal ganglia of the rat (Molleret al., 1993). In similarity to rodents, only a small fraction of porcinesmall-sized skin-projecting neurons expressed GAL (mouse -Brumovsky et al., 2006; rat - Kashiba et al., 1992).

Present results showed that an overwhelming majority of small-sized FB+/SP+ neurons were simultaneously CGRP- and/or PACAP-IRand that N20% of them co-expressed GAL. The pattern of co-expressionof medium-sized FB+/SP+ cells was similar but only a half of these cellsco-expressed PACAP. In contrast, large-sized FB+/SP+ neurons co-expressed CGRP and/or GAL sporadically aswell as theywere complete-ly devoid of PACAP. This picture in the pig correspondswell with resultsof previous studies in human (Landry et al., 2003), rat (Louis et al., 1989;Moller et al., 1993), mouse (Hsieh et al., 2012; Usoskin et al., 2015),guinea pig (Gibbins et al., 1985) and goat (Shin et al., 2003). It also ap-pears to be worth noting, that the relative numbers of FB+/SP+/CGRP+ perikarya projecting either to the urinary bladder trigone(Russo et al., 2013) or to the hindlimb skin of the pig (present study)were similar, and that such coded small-sized cells represent theviscero- or somatosensory nociceptors, respectively. Such an assump-tion is further substantiated in a study by Djouhri et al. (2003), in


which the presence of sensory neuron-specific sodium channelNav1.8, a marker of nociceptors, was negatively correlated with thesoma size, thus once again corroborating the small-sized cells asnociceptors. As may be judged form the available literature, also GALand PACAP should be regarded as putative transmitters in at least a frac-tion of nociceptive neurons (Brumovsky et al., 2006; Hökfelt et al., 1987;Kashiba et al., 1992; Mabuchi et al., 2004). In addition to being involvedin noci- and mechanoceptive transmission, SP, CGRP, GAL and PACAP,antidromically released from primary afferent neurons may also con-tribute to the initiation and exacerbation of neurogenic dermatitis(Black, 2002; Brain et al., 1992; Helyes et al., 2015; Louis et al., 1989;Schmidhuber et al., 2008), either by (i) activation of non-neuronalcells to produce and secrete various proinflammatory agents, e.g., inter-leukin-6 (keratinocytes; Shi et al., 2013) or tumor necrosis factor α(mastocytes, endothelial cells; Peters et al., 2006; Roosterman et al.,2006) or by (ii) vasodilatation of cutaneous vessels leading to plasmatransudation and, in turn, edema formation (Brain and Williams,1985; Cao et al., 1999; Kellogg et al., 2010; Warren et al., 1992).

SOM-positive skin-projecting neurons in the rat have previously beenreported as evenly scattered cells in dorsal root (Perry and Lawson, 1998)and trigeminal ganglia (Ambalavanar et al., 2003). In the present study



Fig. 7.Micrographs showing immunohistochemical staining of the Fast Blue (FB)-labeled neurons in the DRG S1–S3· The small-sized arrows showpositive, while large-sized arrows shownegative immunoreactivity of the skin-projecting neurons in theDRG to examined substances. All of the imageswere taken separately from blue (a, b, c, d, e), green (a′, b′, c′, d′, e′) and red(a″, b″, c″, d″, e″) fluorescent channels. a. One small-sized FB+ neuron which were simultaneously a′ SOM+ and a″ GAL+. b. One small-sized FB+ neuron which was simultaneously b′SOM+ and b″ CB+. c. One FB+ small-sized neuron which was simultaneously c′ GAL+ and c″ PACAP−. d. Two small- and medium-sized FB+ neurons which were simultaneously d′NOS+ and d″ PACAP+. e One small-sized FB+ neuronwhichwas simultaneously e′ GAL− and e″ NOS+. Scale bar= 50 μm. (For interpretation of the references to colour in this figurelegend, the reader is referred to the web version of this article.)


we have found that the vast majority of SOM-IR neurons in the porcineDRGs were small- or medium-sized, while only a few of the large-sizedneurons were expressing this peptide (exclusively in lumbar DRGs).Since a fraction of SOM-containing DRG neurons has been shown to co-express SP (cat; Hanesch et al., 1995), CGRP (mouse; Salio and Ferrini,2016) and/or GAL (cat; Garry et al., 1989), it appears that SOM mayalso be involved in nociceptive transmission. However, available dataconcerning the putative involvement of SOM in nociception appear tobe somewhat contradictory, as this peptide has simultaneously been im-plicated as possessing both the pro- (Kuraishi et al., 1985; Morton et al.,1989) and anti-nociceptive properties (Helyes et al., 2000; Taddese etal., 1995). This discrepancy may be probably elucidated by binding ofthe peptide to different subpopulations of receptors within the spinalcord (Carlton et al., 2001). Furthermore, as SOM-IR afferent cells havepreviously been suggested toparticipate in thermal nociceptive transmis-sion in the cat (Morton et al., 1989), a significantly higher number ofSOM-positive perikarya within lumbar DRGs may suggest a leading roleof these ganglia in the conduction of thermal stimuli from the porcinehind leg. On the other hand, as revealed by a recent study of Stantchevaet al. (2016), a subset of SOM-positive cells in mice belongs to a


subpopulation of pruriceptors, thus suggesting that a fraction of porcineSOM-IR skin-projecting perikarya may also conduct this sensation. How-ever, both these hypotheses need to be verified in detail. In addition tothe conveyance of sensory modalities, SOM has also be implicated totake part in the regulation of both the neurogenic andnon-neurogenic in-flammatory reactions (Helyes et al., 2001; Szolcsányi et al., 1998) at leastpartly by interferingwith the release of SP from peripheral sensory nerveendings (Lembeck et al., 1982).

The inter- as well as intra-ganglionic distribution of porcine skin-projecting nNOS-positive neurons corresponds well with availabledata (rat; Kim et al., 2011, Thippeswamy and Morris, 2002). It isworth mentioning, however, that in the pig, the hindlimb skin-projecting medium-sized nNOS-positive neurons were significantlymore numerous in sacral DRGs, which may suggest that these gangliamay eventually be considered as the anatomical substrate for pain hy-persensitivity evoked by various pathological conditions affecting theskin of the leg (Kim et al., 2011).

As revealed by the present study, a large subset of small-sized FB+/PACAP+ neurons co-expressed GAL and/or nNOS, while medium-sizedFB+/PACAP+ cells often co-expressed nNOS but only a small fraction




of them co-expressed GAL. The large-sized FB+/PACAP+ neurons wereusually devoid of GAL or nNOS. As of now, the co-localization ofPACAP andGAL and/or nNOS has earlier been reported only in parasym-pathetic neurons of rat sphenopalatine ganglion (Csati et al., 2012). Tak-ing into consideration the eminent role of PACAP- and/or GAL- or nNOS-containing afferent perikarya in the regulation of pain processing (Kimet al., 2011; Mabuchi et al., 2003, 2004; Wang et al., 2013), inflamma-tion-driven mechanisms (Helyes et al., 2015; Hökfelt et al., 1987;Kellogg et al., 2010; Schmidhuber et al., 2008; Warren et al., 1992)and/or regulation of the cutaneous blood flow (Kellogg et al.,1998;Shastry et al., 1998), it appears reasonable to hypothesize, that the dif-ferences in the allocation pattern of nNOS-expressing sensory neuronswithin the lumbar and sacral DRGs may reflect a “target-oriented” dis-tribution pattern of such coded perikarya, with lumbar ganglia beingto a greater extent “responsible” for conveying the somatosensory mo-dalities,while the sacral ganglia beingmore profoundly involved in con-trol of viscera functioning. However, this hypothesis needs to be furtherverified in detail.

Similarly, as described previously in the rat (Duc et al., 1994;Honda, 1995), and also in the present study, only a few small-sizedCB-IR skin-projecting neurons were observed. Interestingly, it hasbeen found that the subpopulation of CB-IR neurons was significant-ly more numerous in lumbar, than in sacral DRGs; the potentialphysiological relevance of this observation remains, however, ob-scure as of yet. Although it appears possible that these cells may rep-resent a subset of rapidly-adapting LTMRs, as it has been previouslyproposed by Duc et al. (1994), this hypothesis needs to be furtherelucidated.

As revealed by the labeling of FB+ cells by a mixture made of all anti-bodies used in the present study, it has been demonstrated that a sub-stantial fraction of large-sized neurons, located in sacral DRGs, wasdevoid of any studied neuropeptides, while all of them were expressingneurofilament 200 (NF200; unpublished data). As NF200 immunoreac-tivity is widely accepted as a marker of large myelinated A-β fiber neu-rons (Ruscheweyh et al., 2007), it appears that this subset of skin-projecting DRG cells represented light-touchmechanoreceptors (for a re-view see Abraira and Ginty, 2013); however, the physiological relevanceof their exclusive presence in the sacral DRGs remains obscure as of yet.

In conclusion, the present study for the first time provides a detaileddescription of differences in both the inter- as well as the intra-gangli-onic distribution pattern and the chemical characteristics of porcineskin-projecting DRG neurons, implying their putatively different func-tions. However, further studies utilizing more suitable neuronalmarkers (Le Pichon and Chesler, 2014) or using the transcriptome re-search-based approach, as it has recently been performed in mice (Liet al., 2016; Usoskin et al., 2015), have to be carried out in order to elu-cidate the physiological relevance of each subpopulation of neurons in-nervating the skin of the porcine hindlimb in detail.

Acknowledgments

This studywas supported by the statutory grant No. 25.610.001-300,Faculty of Medical Sciences, the University of Warmia and Mazury inOlsztyn, Poland.

References

Abraira, V.E., Ginty, D.D., 2013. The sensory neurons of touch. Neuron 79 (4):618–639.http://dx.doi.org/10.1016/j.neuron.2013.07.051.

Ambalavanar, R., Moritani, M., Haines, A., Hilton, T., Dessem, D., 2003. Chemical pheno-types of muscle and cutaneous afferent neurons in the rat trigeminal ganglion.J. Comp. Neurol. 460 (2):167–179. http://dx.doi.org/10.1002/cne.10655.

Aoki, Y., Ohtori, S., Takahashi, K., Ino, H., Douya, H., Ozawa, T., Saito, T., Moriya, H., 2005.Expression and co-expression of VR1, CGRP, and IB4-binding glycoprotein in dorsalroot ganglion neurons in rats: differences between the disc afferents and the cutane-ous afferents. Spine (Phila Pa 1976) 1; 30 (13), 1496–1500.

Avon, S.L., Wood, R.E., 2005. Porcine skin as an in-vivo model for ageing of human bitemarks. J Forensic Odontostomatol. 23 (2), 30–39.


Barabas, M.E., Stucky, C.L., 2013. TRPV1, but not TRPA1, in primary sensory neurons con-tributes to cutaneous incision-mediated hypersensitivity. Mol. Pain 9 (9). http://dx.doi.org/10.1186/1744-8069-9-9.

Black, P.H., 2002. Stress and the inflammatory response: a review of neurogenic inflam-mation. Brain Behav. Immun. 16, 22–653.

Blais, M., Mottier, L., Germain, M.A., Bellenfant, S., Cadau, S., Berthod, F., 2014. Sensoryneurons accelerate skin reepithelialization via substance P in an innervated tissue-engineered wound healing model. Tissue Eng. Part A 20 (15–16):2180–2188.http://dx.doi.org/10.1089/ten.tea.2013.0535.

Bossowska, A., Crayton, R., Radziszewski, P., Kmiec, Z., Majewski, M., 2009. Distributionand neurochemical characterization of sensory dorsal root ganglia neurons supplyingporcine urinary bladder. J. Physiol. Pharmacol. (Suppl. 4), 77–81.

Brain, S.D., Williams, T.J., 1985. Inflammatory oedema induced by synergism between cal-citonin gene-related peptide (CGRP) and mediators of increased vascular permeabil-ity. Br. J. Pharmacol. 86 (4), 855–860.

Brain, S.D., Cambridge, H., Hughes, S.R., Wilsoncroft, P., 1992. Evidence that calcitoningene-related peptide contributes to inflammation in the skin and joint. Ann. N. Y.Acad. Sci. 657, 412–419.

Brumovsky, P., Hygge-Blakeman, K., Villar, M.J., Watanabe, M., Wiesenfeld-Hallin, Z.,Hökfelt, T., 2006. Phenotyping of sensory and sympathetic ganglion neurons of agalanin-overexpressing mouse-possible implications for pain processing. J. Chem.Neuroanat. (4):243–262 http://dx.doi.org/10.1016/j.jchemneu.2006.02.001.

Byers, M.R., Bonica, J.J., 2001. Peripheral pain mechanisms and nociceptor plasticity. In:Loeser, L.D., Butler, S.H., Chapman, C.R., Turk, D.C. (Eds.), Bonica's Management ofPain .

Cao, T., Gerard, N.P., Brain, S.D., 1999. Use of NK(1) knockout mice to analyze substanceP-induced edema formation. Am. J. Phys. 277 (2 Pt 2), R476–R481.

Carlton, S.M., Du, J., Zhou, S., Coggeshall, R.E., 2001. Tonic control of peripheral cutaneousnociceptors by somatostatin receptors. J. Neurosci. 21, 4042–4049.

Csati, A., Tajti, J., Kuris, A., Tuka, B., Edvinsson, L., Warfvinge, K., 2012. Distribution of va-soactive intestinal peptide, pituitary adenylate cyclase-activating peptide, nitricoxide synthase, and their receptors in human and rat sphenopalatine ganglion. Neu-roscience 202:158–168. http://dx.doi.org/10.1016/j.neuroscience.2011.10.055.

Di Giminiani, P., Petersen, L.J., Herskin, M.S., 2013. Nociceptive responses to thermal andmechanical stimulations in awake pigs. Eur. J. Pain 17:638–648. http://dx.doi.org/10.1002/j.1532-2149.2012.00228.x.

Djouhri, L., Fang, X., Okuse, K., Wood, J.N., Berry, C.M., Lawson, S., 2003. The TTX-resistantsodium channel Nav1.8 (SNS/PN3): expression and correlation with membraneproperties in rat nociceptive primary afferent neurons. J. Physiol. Lond. 550:739–752. http://dx.doi.org/10.1113/jphysiol.2003.042127.

Duc, C., Barakat-Walter, I., Droz, B., 1994. Innervation of putative rapidly adapting mech-anoreceptors by calbindin- and calretinin-immunoreactive primary sensory neuronsin the rat. Eur. J. Neurosci. 6 (2), 264–271.

Garry, M.G., Miller, K.E., Seybold, V.S., 1989. Lumbar dorsal root ganglia of the cat: a quan-titative study of peptide immunoreactivity and cell size. J. Comp. Neurol. 284 (1),36–47.

Gibbins, I.L., Furness, J.B., Costa, M., Macintyre, I., Hillyard, C.J., Girgis, S., 1985. Co-localizationof calcitonin gene-related peptide like immunoreactivity with substance P in cutaneousvascular and visceral sensory neurons of guinea-pigs. Neurosci. Lett. 57, 125–130.

Gilbert, S.F., 2000. Developmental Biology. sixth ed. Sinauer Associates, Sunderland (MA) .Granstein, R.D., Wagner, J.A., Stohl, L.L., Ding, W., 2015. Calcitonin gene-related peptide:

key regulator of cutaneous immunity. Acta Physiol (Oxford) 213 (3):586–594.http://dx.doi.org/10.1111/apha.12442.

Hanesch, U., Heppelmann, B., Schmidt, R.F., 1995. Somatostatin-like immunoreactivity inprimary afferents of the medial articular nerve and colocalization with substance P inthe cat. J. Comp. Neurol. 354 (3):345–352. http://dx.doi.org/10.1002/cne.903540304.

Harper, A.A., Lawson, S.N., 1985. Electrical properties of rat dorsal root ganglion neuroneswith different peripheral nerve conduction velocities. J. Physiol. 359, 47–63.

Helyes, Z., Thán, M., Oroszi, G., Pintér, E., Németh, J., Kéri, G., Szolcsányi, J., 2000.Anti-nociceptive effect induced by somatostatin released from sensory nerveterminals and by synthetic somatostatin analogues in the rat. Neurosci. Lett.278 (3), 185–188.

Helyes, Z., Pintér, E., Németh, J., Kéri, G., Thán, M., Oroszi, G., Horváth, A., Szolcsányi, J.,2001. Anti-inflammatory effect of synthetic somatostatin analogues in the rat. Br.J. Pharmacol. 134 (7):1571–1579. http://dx.doi.org/10.1038/sj.bjp.0704396.

Helyes, Z., Kun, J., Dobrosi, N., Sándor, K., Németh, J., Perkecz, A., Pintér, E., Szabadfi, K.,Gaszner, B., Tékus, V., Szolcsányi, J., Steinhoff, M., Hashimoto, H., Reglődi, D., Bíró, T.,2015. Pituitary adenylate cyclase-activating polypeptide is upregulated in murineskin inflammation and mediates transient receptor potential Vanilloid-1-inducedneurogenic edema. J. Invest. Dermatol. 135 (9):2209–2218. http://dx.doi.org/10.1038/jid.2015.156.

Herron, A.J., 2009. Pigs as Dermatologic Models of Human Skin Disease. Proceedingof the ACVP/ASVCP Concurrent Annual Meetings, December 5–9, 2009, Monterey.California, USA .

Hill, C.E., Harrison, B.J., Rau, K.K., Hougland, M.T., Bunge, M.B., Mendell, L.M., Petruska, J.C.,2010. Skin incision induces expression of axonal regeneration-related genes in adultrat spinal sensory neurons. J. Pain 11 (11):1066–1073. http://dx.doi.org/10.1016/j.jpain.2010.02.001.

Hökfelt, T.,Wiesenfeld-Hallin, Z., Villar, M., Melander, T., 1987. Increase of galanin-like im-munoreactivity in rat dorsal root ganglion cells after peripheral axotomy. Neurosci.Lett. 83 (3), 217–220.

Honda, C.N., 1995. Differential distribution of calbindin-D28k and parvalbumin in somaticand visceral sensory neurons. Neuroscience 68 (3), 883–892.

Hruban, V., Horak, V., Fortyn, K., Hradecky, J., Klaudy, J., Smith, D.M., Reisnerova, H.,Majzlik, I., 2004. Inheritance of malignantmelanoma in theMeLiM strain ofminiaturepigs. Vet. Med. Czech. 49, 453–459.


http://dx.doi.org/10.1016/j.neuron.2013.07.051

http://dx.doi.org/10.1002/cne.10655

http://refhub.elsevier.com/S0143-4179(16)30011-7/rf0015





http://dx.doi.org/10.1186/1744-8069-9-9



http://dx.doi.org/10.1089/ten.tea.2013.0535










http://dx.doi.org/10.1016/j.jchemneu.2006.02.001








http://dx.doi.org/10.1016/j.neuroscience.2011.10.055

http://dx.doi.org/10.1002/j.1532-2149.2012.00228.x

http://dx.doi.org/10.1113/jphysiol.2003.042127











http://dx.doi.org/10.1111/apha.12442







http://dx.doi.org/10.1038/sj.bjp.0704396

http://dx.doi.org/10.1038/jid.2015.156

http://dx.doi.org/10.1038/jid.2015.156




http://dx.doi.org/10.1016/j.jpain.2010.02.001

http://dx.doi.org/10.1016/j.jpain.2010.02.001










Hsieh, Y.L., Lin, C.L., Chiang, H., Fu, Y.S., Lue, J.H., Hsieh, S.T., 2012. Role of peptidergic nerveterminals in the skin: reversal of thermal sensation by calcitonin gene-related pep-tide in TRPV1-depleted neuropathy. PLoS One 7 (11), e50805. http://dx.doi.org/10.1371/journal.pone.0050805.

Johnson, M.S., 2008. The Role of Cutaneous Innervation in the Sensory Abnormalities As-sociated with Diabetic Neuropathy. University of Kansas (Dissertation).

Kashiba, H., Senba, E., Ueda, Y., Tohyama, M., 1990. Calbindin D28k-containing splanchnicand cutaneous dorsal root ganglion neurons of the rat. Brain Res. 528 (2), 311–316.

Kashiba, H., Senba, E., Ueda, Y., Tohyama, M., 1992. Co-localized but target-unrelated ex-pression of vasoactive intestinal polypeptide and galanin in rat dorsal root ganglionneurons after peripheral nerve crush injury. Brain Res. 582 (1), 47–57.

Kashiba, H., Noguchi, K., Ueda, Y., Senba, E., 1994. Neuropeptide Y and galanin arecoexpressed in rat large type A sensory neurons after peripheral transection. Peptides15 (3), 411–416.

Kellogg Jr., D.L., Crandall, C.G., Liu, Y., Charkoudian, N., Johnson, J.M., 1998. Nitric oxide andcutaneous active vasodilation during heat stress in humans. J. Appl. Physiol. 85 (3),824–829.

Kellogg Jr., D.L., Zhao, J.L., Wu, Y., Johnson, J.M., 2010. VIP/PACAP receptor mediation of cu-taneous active vasodilation during heat stress in humans. J. Appl. Physiol. 109 (1):95–100. http://dx.doi.org/10.1152/japplphysiol.01187.2009.

Kim, K.H., Kim, J.I., Han, J.A., Choe, M.A., Ahn, J.H., 2011. Upregulation of neuronal nitricoxide synthase in the periphery promotes pain hypersensitivity after peripheralnerve injury. Neuroscience 190:367–378. http://dx.doi.org/10.1016/j.neuroscience.2011.05.064.

Kim, J.W., Lee, D.W., Choi, W.H., Jeon, Y.R., Kim, S.H., Cho, H., Lee, E.J., Hong, Z.Y., Lee, W.J.,Cho, J., 2013. Development of a porcine skin injury model and characterization of thedose-dependent response to high-dose radiation. J. Radiat. Res. 54 (5):823–831.http://dx.doi.org/10.1093/jrr/rrt016.

Kuraishi, Y., Hirota, N., Sato, Y., Hino, Y., Satoh, M., Takagi, H., 1985. Evidence that sub-stance P and somatostatin transmit separate information related to pain in the spinaldorsal horn. Brain Res. 325 (1–2), 294–298.

Landry, M., Aman, K., Dostrovsky, J., Lozano, A.M., Carlstedt, T., Spenger, C., Josephson, A.,Wiesenfeld-Hallin, Z., Hökfelt, T., 2003. Galanin expression in adult human dorsalroot ganglion neurons: initial observations. Neuroscience 117 (4), 795–809.

Le Pichon, C.E., Chesler, A.T., 2014. The functional and anatomical dissection of somato-sensory subpopulations using mouse genetics. Front. Neuroanat. 8:21. http://dx.doi.org/10.3389/fnana.2014.00021.

Lembeck, F., Donnerer, J., Barthó, L., 1982. Inhibition of neurogenic vasodilation and plas-ma extravasation by substance P antagonists, somatostatin and [D-Met2, Pro5]enkephalinamide. Eur. J. Pharmacol. 85 (2), 171–176.

Levi, J.R., Veerappan, A., Chen, B., Mirkov, M., Sierra, R., Spiegel, J.H., 2011. Histologic eval-uation of laser lipolysis comparing continuous wave vs pulsed lasers in an in vivo pigmodel. Arch. Facial Plast. Surg. 13 (1):41–50. http://dx.doi.org/10.1001/archfacial.2010.103.

Li, C.L., Li, K.C., Wu, D., Chen, Y., Luo, H., Zhao, J.R., Wang, S.S., Sun, M.M., Lu, Y.J., Zhong,Y.Q., Hu, X.Y., Hou, R., Zhou, B.B., Bao, L., Xiao, H.S., Zhang, X., 2016.Somatosensory neuron types identified by high-coverage single-cell RNA-sequencingand functional heterogeneity. Cell Res. 26 (1):83–102. http://dx.doi.org/10.1038/cr.2015.149.

Louis, S.M., Jamieson, A., Russell, N.J., Dockray, G.J., 1989. The role of substance P and cal-citonin gene-related peptide in neurogenic plasma extravasation and vasodilatationin the rat. Neuroscience 32 (3), 581–586.

Ma, Q., 2010. Labeled lines meet and talk: population coding of somatic sensations. J. Clin.Invest. 120 (11):3773–3778. http://dx.doi.org/10.1172/JCI43426.

Mabuchi, T., Matsumura, S., Okuda-Ashitaka, E., Kitano, T., Kojima, H., Nagano, T., Minami,T., Ito, S., 2003. Attenuation of neuropathic pain by the nociceptin/orphanin FQ antag-onist JTC-801 is mediated by inhibition of nitric oxide production. Eur. J. Neurosci. 17(7), 1384–1392.

Mabuchi, T., Shintani, N., Matsumura, S., Okuda-Ashitaka, E., Hashimoto, H., Muratani, T.,Minami, T., Baba, A., Ito, S., 2004. Pituitary adenylate cyclase-activating polypeptide isrequired for the development of spinal sensitization and induction of neuropathicpain. J. Neurosci. 24 (33):7283–7291. http://dx.doi.org/10.1523/JNEUROSCI.0983-04.2004.

McGlone, F., Reilly, D., 2010. The cutaneous sensory system. Neurosci. Biobehav. Rev. 34(2):148–159. http://dx.doi.org/10.1016/j.neubiorev.2009.08.004.

McIlvain, H.B., Baudy, A., Sullivan, K., Liu, D., Pong, K., Fennell, M., Dunlop, J., 2006. Pitui-tary adenylate cyclase-activating peptide (PACAP) induces differentiation in the neu-ronal F11 cell line through a PKA-dependent pathway. Brain Res. 1077 (1):16–23.http://dx.doi.org/10.1016/j.brainres.2005.12.130.

McLachlan, E.M., Jänig, W., 1983. The cell bodies of origin of sympathetic and sensoryaxons in some skin and muscle nerves of the cat hindlimb. J. Comp. Neurol. 214(2):115–130. http://dx.doi.org/10.1002/cne.902140202.

Minett, M.S., Eijkelkamp, N., Wood, J.N., 2014. Significant determinants of mouse pain be-haviour. PLoS One 9 (8), e104458. http://dx.doi.org/10.1371/journal.pone.0104458.

Molander, C., Ygge, J., Dalsgaard, C.J., 1987. Substance P-, somatostatin- and calcitoningene-related peptide-like immunoreactivity and fluoride resistant acid phospha-tase-activity in relation to retrogradely labeled cutaneous, muscular and visceral pri-mary sensory neurons in the rat. Neurosci. Lett. 74 (1), 37–42.

Moller, K., Zhang, Y.Z., Hakanson, R., Luts, A., Sjolund, B., Uddman, R., Sundler, F., 1993. Pi-tuitary adenylate cyclase activating peptide is a sensory neuropeptide: immunocyto-chemical and immunochemical evidence. Neuroscience 57, 725–732.

Morton, C.R., Hutchison, W.D., Hendry, I.A., Duggan, A.W., 1989. Somatostatin: evidencefor a role in thermal nociception. Brain Res. 488 (1–2), 89–96.

Olivry, T., Mirsky, M.L., Singleton, W., Dunston, S.M., Borrillo, A.K.G., Xu, L., Traczyk, T.,Rosolia, D.L., Chan, L.S., 2000. A spontaneously arising porcine model of bullous pem-phigoid. Arch. Dermatol. Res. 292 (1), 37–45.


Perry, M.J., Lawson, S.N., 1998. Differences in expression of oligosaccharides, neuro-peptides, carbonic anhydrase and neurofilament in rat primary afferent neuronsretrogradely labeled via skin, muscle or visceral nerves. Neuroscience 85 (1),293–310.

Peters, E.M., Ericson, M.E., Hosoi, J., Seiffert, K., Hordinsky, M.K., Ansel, J.C., Paus, R.,Scholzen, T.E., 2006. Neuropeptide control mechanisms in cutaneous biology: physi-ological and clinical significance. J. Invest. Dermatol. 126 (9), 1937–1947.

Petruska, J.C., Napaporn, J., Johnson, R.D., Gu, J.G., Cooper, B.Y., 2000. Subclassified acutelydissociated cells of rat DRG: histochemistry and patterns of capsaicin-, proton-, andATP-activated currents. J. Neurophysiol. 84, 2365–2379.

Plenderleith, M.B., Snow, P.J., 1993. The plant lectin Bandeiraeasimplicifolia I-B4 identifiesa subpopulation of small diameter primary sensory neurones which innervate theskin in the rat. Neurosci. Lett. 159 (1–2), 17–20.

Roosterman, D., Goerge, T., Schneider, S.W., Bunnett, N.W., Steinhoff, M., 2006. Neuronalcontrol of skin function: the skin as a neuroimmunoendocrine organ. Physiol. Rev.86 (4), 1309–1379.

Rowe, M., Iwamura, Y., 2001. Somatosensory Processing: From Single Neuron to Brain Im-aging. Harwood Academic, Newark, NJ .

Royden, H.J., Burns, T.M., Aminoff, M.J., Scott, L., 2013. The netter collection of medical il-lustrations: nervous system. Volume 7.

Ruscheweyh, R., Forsthuber, L., Schoffnegger, D., Sandkühler, J., 2007. Modification of clas-sical neurochemical markers in identified primary afferent neurons with Abeta-,Adelta-, and C-fibers after chronic constriction injury in mice. J. Comp. Neurol. 502(2), 325–336.

Russo, D., Clavenzani, P., Sorteni, C., Bo Minelli, L., Botti, M., Gazza, F., Panu, R., Ragionieri,L., Chiocchetti, R., 2013. Neurochemical features of boar lumbosacral dorsal root gan-glion neurons and characterization of sensory neurons innervating the urinary blad-der trigone. J. Comp. Neurol. 521 (2):342–366. http://dx.doi.org/10.1002/cne.23177.

Salio, C., Ferrini, F., 2016. BDNF and GDNF expression in discrete populations ofnociceptors. Ann. Anat. 2016 (207):55–61. http://dx.doi.org/10.1016/j.aanat.2015.12.001.

Schmidhuber, S.M., Starr, A., Wynick, D., Kofler, B., Brain, S.D., 2008. Targeted disruption ofthe galanin gene attenuates inflammatory responses in murine skin. J. Mol. Neurosci.34 (2):149–155. http://dx.doi.org/10.1007/s12031-007-9015-9.

Shastry, S., Dietz, N.M., Halliwill, J.R., Reed, A.S., Joyner, M.J., 1998. Effects of nitric oxidesynthase inhibition on cutaneous vasodilation during body heating in humans.J. Appl. Physiol. 85 (3), 830–834 (1985).

Shi, X., Wang, L., Clark, J.D., Kingery, W.S., 2013. Keratinocytes express cytokines andnerve growth factor in response to neuropeptide activation of the ERK1/2 and JNKMAPK transcription pathways. Regul. Pept. 186:92–103. http://dx.doi.org/10.1016/j.regpep.2013.08.001.

Shi, T.J., Xiang, Q., Zhang, M.D., Barde, S., Kai-Larsen, Y., Fried, K., Josephson, A., Glück, L.,Deyev, S.M., Zvyagin, A.V., Schulz, S., Hökfelt, T., 2014. Somatostatin and its 2A receptorin dorsal root ganglia and dorsal horn of mouse and human: expression, trafficking andpossible role in pain. Mol. Pain 10:12. http://dx.doi.org/10.1186/1744-8069-10-12.

Shin, N.S., Lee, I.S., Yoon, Y.S., Lee, H.S., 2003. Immunohistochemical localization of substanceP, calcitonin gene-related peptide, galanin and calcium-binding proteins in trigeminalganglia of goat (Capra hircus). Anat. Histol. Embryol. 32 (5), 310–315.

Smith, E.S., Lewin, G.R., 2009. Nociceptors: a phylogenetic view. J. Comp. Physiol ANeuroethol. Sens. Neural Behav. Physiol. 195:1089–1106. http://dx.doi.org/10.1007/s00359-009-0482-z.

Sobociński, M., 1973. Nervous System of Domestic Animals. fourth ed. State ScientificPublishing House .

Stantcheva, K.K., Iovino, L., Dhandapani, R., Martinez, C., Castaldi, L., Nocchi, L., Perlas, E.,Portulano, C., Pesaresi, M., Shirlekar, K.S., de Castro Reis, F., Paparountas, T., Bilbao,D., Heppenstall, P.A., 2016. A subpopulation of itch-sensing neurons marked by Retand somatostatin expression. EMBO Rep. 17 (4):585–600. http://dx.doi.org/10.15252/embr.201540983.

Sullivan, T.P., Eaglstein, W.H., Davis, S.C., Mertz, P., 2001. The pig as a model for humanwound healing. Wound Repair Regen. 9, 66–76.

Suzuki, H., Aoyama, Y., Senzaki, K., Vincler, M., Wittenauer, S., Yoshikawa, M., Ozaki, S.,Oppenheim, R.W., Shiga, T., 2010. Characterization of sensory neurons in the dorsalroot ganglia of Bax-deficient mice. Brain Res. 1362:23–31. http://dx.doi.org/10.1016/j.brainres.2010.09.027.

Szolcsányi, J., Pintér, E., Helyes, Z., Oroszi, G., Németh, J., 1998. Systemic anti-inflammatoryeffect induced by counter-irritation through a local release of somatostatin fromnociceptors. Br. J. Pharmacol. 125 (4), 916–922.

Taddese, A., Nah, S.Y., Mc Cleskey, E.W., 1995. Selective opioid inhibition of small nocicep-tive neurons. Science 270, 1366–1369.

Takahashi, A., Wada, Y., Ohtori, S., Saisu, T., Moriya, H., 2003. Application of shock wavesto rat skin decreases calcitonin gene-related peptide immunoreactivity in dorsal rootganglion neurons. Auton. Neurosci. 107 (2):81–84. http://dx.doi.org/10.1016/S1566-0702(03)00134-6.

Thippeswamy, T., Morris, R., 2002. The roles of nitric oxide in dorsal root ganglion neu-rons. Ann. N. Y. Acad. Sci. 962, 103–110.

Usoskin, D., Furlan, A., Islam, S., Abdo, H., Lönnerberg, P., Lou, D., Hjerling-Leffler, J.,Haeggström, J., Kharchenko, O., Kharchenko, P.V., Linnarsson, S., Ernfors, P., 2015. Un-biased classification of sensory neuron types by large-scale single-cell RNA sequenc-ing. Nat. Neurosci. 18 (1):145–153. http://dx.doi.org/10.1038/nn.3881.

Vana, G., Meingassner, J.G., 2000. Morphologic and immunohistochemical features of ex-perimentally induced allergic contact dermatitis in Göttingen mini pigs. Vet. Pathol.37 (6), 565–580.

Wang, H.Y., Tsai, Y.J., Chen, S.H., Lin, C.T., Lue, J.H., 2013. Lysophosphatidylcholine causesneuropathic pain via the increase of neuronal nitric oxide synthase in the dorsalroot ganglion and cuneate nucleus. Pharmacol. Biochem. Behav. 106:47–56. http://dx.doi.org/10.1016/j.pbb.2013.03.002.


http://dx.doi.org/10.1371/journal.pone.0050805















http://dx.doi.org/10.1152/japplphysiol.01187.2009



http://dx.doi.org/10.1093/jrr/rrt016






http://dx.doi.org/10.3389/fnana.2014.00021




http://dx.doi.org/10.1001/archfacial.2010.103

http://dx.doi.org/10.1001/archfacial.2010.103

http://dx.doi.org/10.1038/cr.2015.149

http://dx.doi.org/10.1038/cr.2015.149




http://dx.doi.org/10.1172/JCI43426




http://dx.doi.org/10.1523/JNEUROSCI.0983-04.2004

http://dx.doi.org/10.1523/JNEUROSCI.0983-04.2004

http://dx.doi.org/10.1016/j.neubiorev.2009.08.004

http://dx.doi.org/10.1016/j.brainres.2005.12.130






































http://dx.doi.org/10.1016/j.aanat.2015.12.001

http://dx.doi.org/10.1016/j.aanat.2015.12.001

http://dx.doi.org/10.1007/s12031-007-9015-9




http://dx.doi.org/10.1016/j.regpep.2013.08.001

http://dx.doi.org/10.1016/j.regpep.2013.08.001

http://dx.doi.org/10.1186/1744-8069-10-12




http://dx.doi.org/10.1007/s00359-009-0482-z

http://dx.doi.org/10.1007/s00359-009-0482-z



http://dx.doi.org/10.15252/embr.201540983

http://dx.doi.org/10.15252/embr.201540983










http://dx.doi.org/10.1016/S1566-0702(03)00134-6

http://dx.doi.org/10.1016/S1566-0702(03)00134-6



http://dx.doi.org/10.1038/nn.3881




http://dx.doi.org/10.1016/j.pbb.2013.03.002



Warren, J.B., Larkin, S.W., Coughlan, M., Kajekar, R., Williams, T.J., 1992. Pituitary adenyl-ate cyclase activating polypeptide is a potent vasodilator and oedema potentiator inrabbit skin in vivo. Br. J. Pharmacol. 106 (2), 331–334.

Wessels, W.J., Feirabend, H.K., Marani, E., 1990. Somatotopic organization in the sensoryinnervation of the rat hindlimb during development, using half dorsal root gangliaas subsegmental units. Eur. J. Morphol. 28 (2–4), 394–403.

Ye, Y., Bae, S.S., Viet, C.T., Troob, S., Bernabé, D., Schmidt, B.L., 2013. IB4(+) and TRPV1(+) sen-sory neurons mediate pain but not proliferation in a mouse model of squamous cell car-cinoma. Pharmacol. Biochem. Behav. 106. http://dx.doi.org/10.1186/1744-9081-10-5.


Yeomans, D.C., Jones, T., Laurito, C.E., Lu, Y., Wilson, S.P., 2004. Reversal of ongoing thermalhyperalgesia in mice by a recombinant herpesvirus that encodes humanpreproenkephalin. Mol. Ther. 9:24–29. http://dx.doi.org/10.1016/j.ymthe.2003.10.008.

Zhu, K.Q., Engrav, L.H., Gibran, N.S., Cole, J.K., Matsumura, H., Piepkorn, M., Isik, F.F.,Carrougher, G.J., Muangman, P.M., Yunusov, M.Y., Yang, T.M., 2003. The female, redDuroc pig as an animal model of hypertrophic scarring and the potential role of thecones of skin. Burns 29, 649–664.

Zimmerman, A., Bai, L., Ginty, D.D., 2014. The gentle touch receptors of mammalian skin.Science 346 (6212):950–954. http://dx.doi.org/10.1126/science.1254229.








http://dx.doi.org/10.1186/1744-9081-10-5

http://dx.doi.org/10.1016/j.ymthe.2003.10.008




http://dx.doi.org/10.1126/science.1254229


distribution and chemical coding of sensory neurons innervating … · 2016. 12. 20. ·...

Documents