All experimental procedures were approved by the Institutional Animal Care and Use Committee of Shiga University of Medicine, and were designed to minimize the number of animals and their suffering in accordance with the 1996 NIH Guide for the Care and Use of Laboratory Animals.

Nine adult male Wistar rats (Clea Japan, Tokyo, Japan) weighing 200–300 g were used. They were deeply anesthetized by an intraperitoneal injection of sodium pentobarbital (150 mg/kg), and transcardially perfused with 10 mM phosphate-buffered saline (PBS, pH 7.4) at 20°C. For reverse transcription-polymerase chain reaction (RT-PCR) and Western blot analyses, spinal cord segments and dorsal root ganglion (DRG) samples were removed and stored at −80°C. For immunohistochemical studies, rats were similarly anesthetized, perfused transcardially with PBS, and further perfused with an ice-cold fixative of 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). Spinal cord segments were collected and post-fixed by immersion for 24 h in the same fixative at 4°C. After cryoprotection for 4 days in phosphate buffer containing 15% sucrose at 4°C, 20 μm sections were cut in a cryostat. These sections were collected and stored in 0.1 M PBS containing 0.3% Triton X-100 (PBST) at 4°C.

Total RNA was extracted from DRG and spinal cord samples using a Fastpure RNA kit (Takara Bio, Otsu, Japan) and treated with DNAse (Turbo DNAse, Ambion, Austin, TX, USA) prior to reverse transcription. One microgram of total RNA was reverse-transcribed into cDNA using SuperScript III (Thermo Fisher Scientific Inc., Waltham, MA, USA) and an oligo dT adapter primer. Specific BNP cDNA was amplified by PCR using Ex Taq HS DNA polymerase (Takara Bio) and the following rat BNP primers: 5′-CAGAAGGTGCTGCCCCAGATG-3′ (sense) and 5′-GACTGCGCCGATCCGGTC-3′ (antisense). The amplification profile consisted of an initial step of denaturation at 95°C for 1 min, 36 cycles of 10 s at 98°C, 30 s at 55°C and 1 min at 72°C, followed by final extension at 72°C for 8 min. The PCR products were electrophoresed on a 2% agarose gel and stained with ethidium bromide.

Spinal cord samples were homogenized in ice-cold RIPA buffer containing protease inhibitors. The homogenates were centrifuged at 15,000 rpm for 20 min at 4°C, and then the supernatants were collected. Protein concentrations were detected using the Bio-Rad Protein Assay kit (Bio-Rad, Hercules, CA, USA). Protein samples (29.4 μg) were electrophoresed on 15% ready-to-use PAGE (WAKO, Japan) and transferred onto PVDF membranes. The membranes were incubated in 20% reagent N102 (NOF Corp., Tokyo, Japan) in TBS for 1 h at room temperature to block non-specific protein binding sites, then incubated overnight at 4°C with a rabbit polyclonal antibody BNP (Millipore, AB1549), at a dilution of 1:2000 in TBS containing 0.1% Tween-20 (TBST) or with normal rabbit serum (1:2000) in TBST as control. The blots were washed three times for 20 min each with TBST, and then incubated for 1 h at room temperature with peroxidase-conjugated goat anti-rabbit IgG (1:25,000; Jackson ImmunoResearch, Baltimore, PA, USA). After three washes with TBST, the blots were developed using Chemi-lumi One Super (Nacalai, Japan), and visualized using a LuminoImager LAS3000 (FujiFilm, Tokyo, Japan).

Serial cryostat sections (20 μm) were floated in 0.1 M PBST (pH 7.4) for 4 days at 4°C and treated in a free-floating state. Endogenous peroxidase activity was blocked by incubating the sections in 0.5% H₂O₂ in PBST for 30 min. After washing the sections three times (10 min each) with PBST, they were blocked in 5% bovine serum albumin (BSA) for 2 h at room temperature. After blocking, they were incubated at 4°C for 3 days with a rabbit polyclonal antibody against rat BNP (1:4000 dilution; Millipore, Temecula, CA, USA). The primary antibody was diluted in 2% BSA in PBST. After the sections had been rinsed three times with PBST (10 min each time), they were incubated for 1 h at room temperature with biotinylated goat anti-rabbit IgG (1:3000 dilution; Vector Laboratories, Burlingame, CA, USA) before being rinsed another three times with PBST. The sections were finally incubated for 1 h at room temperature with an avidin-biotin-peroxidase complex (1:4000 dilution; ABC Elite; Vector Laboratories, Burlingame, CA, USA) and the peroxidase-labeled sections were developed in 0.02% 3,3-diamine-benzidine tetrahydrochloride with 0.07% nickel ammonium sulfate in 50 mM Tris-HCl (pH 7.6), with 0.005% H₂O₂. Negative control sections were subjected to the same procedures without primary antibody.

Free-floating sections of the spinal cord were blocked for 1 h at room temperature with 5% BSA in PBST to block non-specific binding, then incubated for 48 h at 4°C with a mixture of rabbit polyclonal antibody against rat BNP (diluted 1:2000; Millipore) and either guinea-pig polyclonal CGRP antibody (1:200; T-5053, Peninsula Lab, San Carlos, CA, USA) or goat anti-ChAT antibody (diluted 1:500; Millipore). After three rinses with PBST, sections were incubated for 3 h at room temperature with the appropriate secondary antibodies. The secondary antibodies were Alexa Fluor 555-conjugated anti-rabbit IgG, Alexa Fluor 488-conjugated anti-guinea pig IgG, or Alexa Fluor 488-conjugated anti-goat IgG (diluted 1:500, Thermo Fisher Scientific Inc., Waltham, MA, USA). Negative control sections were subjected to the same procedures without primary antibody. After three washes with PBST, the sections were mounted on gelatin-coated glass slides and examined using a confocal laser-scanning microscope (LSM 510 META, Carl Zeiss, Heidelberg, Germany).

The major regions of the spinal cord were identified according to the rat brain atlas (Paxinos and Watson, 2006). Images of selected sections were captured and organized into photographic panels. To improve the visualization of the results, only the contrast and brightness of the images were adjusted using Adobe Photoshop, without any further manipulation of the images.

In this study, we used a primary antibody against rat BNP to examine BNP immunoreactivity in the spinal cord. This antibody was previously characterized (Abdelalim and Tooyama, 2009). In our previous study, we confirmed the specificity of this antibody using Western blotting, where it recognized a single band for BNP in rodent cells (Abdelalim and Tooyama, 2009). Furthermore, in the current study we performed Western blotting to confirm the specificity on rat spinal cord samples. A specific band was detected at approximately 33 kDa in the spinal cord samples, corresponding to BNP protein (Figure 1B). It is known that Western blotting is effective for validating antibody specificity (Burry, 2000) and thus the antibody used in this study can be considered to recognize rat BNP protein specifically.

RT-PCR was used to examine the expression of BNP mRNA in the spinal cord and DRG of rat. This analysis showed a single band for BNP at the expected size (321 bp) in the DRG and spinal cord (Figure 1). Here we included the DRG in the examination because DRG neurons are known to send projections to the dorsal horn (DH) of the spinal cord, which showed BNP immunoreactivity in the current study.

Immunohistochemical analysis revealed BNP immunoreactivity in different regions of the spinal cord (cervical, thoracic, lumbar and sacral) in neuronal cell bodies and nerve fibers (Figure 2). In the negative control sections, no immunoreactivity was observed (data not shown). In the cervical spinal cord segments, BNP-immunoreactive dots were detected throughout the laminae in the gray matter (Figures 2A–C). In the DH, high-intensity of BNP-immunoreactive fibers was noted in the DH (Figures 2A,B). In addition, the immunoreactive dots were evident near the spinal canal (Figures 2A,C). The BNP-immunoreactive cell bodies were detected in the cervical segments and continued throughout the spinal cord (Figures 2A–I). In the DH, BNP immunoreactivity was also seen in nerve fibers (Figures 2D,F). BNP-immunoreactive neurons were clearly observed in the intermediate column (IC) and the ventral horn (VH) of the spinal cord (Figures 2A,C–I). In the VH, strong BNP immunoreactivity was seen in the perikarya and dendrites of the multipolar neurons, indicating its localization in the motor spinal neurons (Figures 2C,E,G,I).

To investigate whether BNP is co-localized with CGRP in sensory and motor systems in the spinal cord, we performed a confocal microscopic analysis of the rat spinal cord sections to recognize CGRP and BNP co-localization. It has been shown that CGRP is expressed in the afferent fibers of laminae I–II of the DH (Eftekhari and Edvinsson, 2011). Double-immunolabeling of BNP and CGRP showed that BNP immunoreactivity was co-localized with CGRP in the DH of the spinal cord, indicating BNP expression in laminae I–II (Figures 3A–C). Furthermore, in the VH, BNP was co-localized with CGRP in the perikarya of the motor neurons (Figures 3D–I).

Furthermore, to examine whether spinal motor (ChAT-positive) neurons express BNP, we performed a laser confocal microscopic analysis of the rat spinal cord to recognize BNP and ChAT co-localization. It is known that ChAT is a marker for motor neurons, which are present in the VH of the spinal cord. BNP immunoreactivity was co-localized with ChAT-positive neurons in the VH of the spinal cord, indicating BNP expression in the rat spinal motor neurons (Figures 4A–F).

In the current study, we detected BNP immunoreactivity in the DH of the spinal cord. Furthermore, a double-immunostaining experiment showed that BNP protein is co-localized with CGRP protein in the DH of the spinal cord. CGRP is released from primary afferent fibers that convey nociceptive information in the spinal cord and plays a crucial role in pain sensitivity (Gangadharan and Kuner, 2013). The expression profile of BNP with CGRP in the DH of the spinal cord suggests that BNP is expressed in the primary afferent terminals of nociceptive DRG neurons. This was confirmed in this study using RT-PCR, where we found that BNP mRNA is also expressed in rat DRG.

Our results agree with earlier studies, in which BNP immunoreactivity was detected in laminae I–II of rat and pig (Kawata et al., 1989; Saper et al., 1989). However, a recent study was able to detect BNP protein in the spinal DH using Western blotting, but could not detect BNP immunoreactivity in the afferent fibers using immunohistochemistry (Zhang et al., 2010). These variations are due to technical differences. In our study, we used very sensitive protocols to detect BNP immunoreactivity. As described above, the rat tissues were fixed using a perfusion system and the immunohistochemical procedure was performed using a free-floating technique, which is more sensitive than other immunohistochemical techniques.

BNP and its receptors (NPR-A and NPR-B) are expressed in rat DRG (Zhang et al., 2010; Abdelalim et al., 2013) and it has been found that their expressions are increased in response to pain induction (Zhang et al., 2010). The BNP protein has been found in CGRP-and isolectin B4 (IB4)-positive DRG neurons (Zhang et al., 2010). Interestingly, our recent study showed the co-localization of NPR-B with CGRP in the same DRG neurons and in the DH of the spinal cord (Abdelalim et al., 2013). It is known that CGRP is expressed in DRG small neurons and afferent fibers in laminae I–II of the spinal cord (Gibson et al., 1984; Schaible, 1996). After acute pain induction in animal models, CGRP level is elevated in the DRG and DH of the spinal cord (Traub et al., 1990; Zhang et al., 1993). Transection of the DH causes a dramatic decrease of the CGRP expression in the laminae I–II in the DH (Fernandez et al., 2003). However, brachial plexus injury leads to an increase in the expression of CGRP in the DH (Chen et al., 2010). These findings indicate the essential role of CGRP in the pain pathway and its level is affected by the nature and type of the nerve injury (Bergman et al., 1996; Li et al., 2004). Taken together, these findings suggest that BNP in the DRG and the spinal cord may be involved in pain modulation through NPR-A and/or NPR-B receptors.

The current study is the first to examine the relationship between BNP and motor neurons in the spinal cord. We found that BNP was co-localized with CGRP in the multipolar neurons in the VH. Our results are consistent with those of the previous studies, which showed that the CGRP is expressed in the motor neuron in the ventral of the spinal cord (Chen et al., 2010; McCoy et al., 2012; Cui et al., 2015; Kim et al., 2016), where it is co-localized with ChAT (Zheng et al., 2008). CGRP plays different roles based on its localization. It has been reported that the expression level of CGRP in the motor neurons changes after nerve injury and these changes depend on the age of the animal and the type of the injury. In the spinal cord, CGRP immunoreactivity is temporarily elevated in laminae I–IV of the DH and motor neurons of the VH after suffering a sciatic nerve injury (Zheng et al., 2008). Another study reported that sciatic nerve injury leads to down-regulation of CGRP transcripts in the motor neurons of the VH in the first week after birth and starts to be unregulated 12 days after birth (Tonra and Mendell, 1998). Recently, it has also been suggested that CGRP expression in the spinal cord is involved in the growth of sensory and motor axons (Kim et al., 2016). Taken together, these findings suggest that BNP might play a role in repair mechanisms following nerve injuries.

We also found that BNP was co-localized with ChAT-positive neurons, indicating its localization in the spinal motor neurons in the VH. Our previous report demonstrated the co-localization of NPR receptors (NPR-A, NPR-B and NPR-C) with ChAT-positive neurons in the central nervous system (Abdelalim et al., 2007, 2008a,b; Abdelalim and Tooyama, 2010). The relationship between NPs and cholinergic (ChAT-positive) neurons has been previously reported. It has been found that stimulation of cholinergic neurons in the anteroventral third ventricle leads to induction of ANP expression hypothalamus (Antunes-Rodrigues et al., 1996). These findings suggest that NPs may play an essential role in mediating motor function by binding to their receptors. These findings also indicate that BNP may be involved in specific motor functions in the spinal cord.