Abstract
The general lack of pain experience is a rare occurrence in humans, and the molecular causes for this phenotype are not well understood. Here we have studied a Canadian family from Newfoundland with members who exhibit a congenital inability to experience pain. We have mapped the locus to a 13.7 Mb region on chromosome 2q (2q24.3–2q31.1). Screening of candidate genes in this region identified a protein-truncating mutation in SCN9A , which encodes for the voltage-gated sodium channel Na v 1.7. The mutation is a C–A transversion at nucleotide 984 transforming the codon for tyrosine 328 to a stop codon. The predicted product lacks all pore-forming regions of Na v 1.7. Indeed, expression of this altered gene in a cell line did not produce functional responses, nor did it cause compensatory effects on endogenous voltage-gated sodium currents when expressed in ND7/23 cells. Because a homozygous knockout of Na v 1.7 in mice has been shown to be lethal, we explored why a deficiency of Na v 1.7 is non-lethal in humans. Expression studies in monkey, human, mouse and rat tissue indicated species-differences in the Na v 1.7 expression profile. Whereas in rodents the channel was strongly expressed in hypothalamic nuclei, only weak mRNA levels were detected in this area in primates. Furthermore, primate pituitary and adrenal glands were devoid of signal, whereas these two glands were mRNA-positive in rodents. This species difference may explain the non-lethality of the observed mutation in humans. Our data further establish Na v 1.7 as a critical element of peripheral nociception in humans.
INTRODUCTION
The accurate perception and rapid response to stimuli that may cause physical damage are essential elements for survival. Not surprisingly then, numerous receptors, ion channels and other proteins are involved in the perception and transmission of nociceptive/painful stimuli. Among them are numerous voltage-gated sodium channels ( 1 ). For example, a knockout of the voltage-gated sodium channel Na v 1.8 in mice has been shown to cause deficits in thermal pain perception ( 2 ). Likewise, mice that have a selective knockout of Na v 1.7 in nociceptive neurons also display deficits in acute and inflammatory pain, whereas a global knockout of Na v 1.7 has been found to be lethal ( 3 ).
In humans, numerous mutations in the voltage-gated sodium channel Na v 1.7 have been shown to be associated with, and likely to be causal for, two rare congenital pain conditions, erythromelalgia ( 4 , 5 ) and paroxysmal extreme pain disorder ( 6 ). In both cases, the mutations cause hyperactivity of the mutated voltage-gated sodium channels, either by shifting the voltage-dependence of activation in the hyperpolarizing direction ( 7 , 8 ) or by attenuating fast inactivation ( 6 ). Furthermore, nonsense mutations of Na v 1.7, producing truncated, non-functional proteins, have recently been described in several families with a congenital inability to experience pain ( 9 , 10 ).
In parallel and independently from the study by Goldberg et al . ( 10 ), we have studied a Canadian family with four members exhibiting a congenital inability to experience pain. We find that the disease phenotype linked through an autosomal recessive mode of inheritance to a mutation in SCN9A, transforming the codon for tyrosine 328 to a stop codon. This is predicted to truncate the voltage-gated sodium channel Na v 1.7 after the fifth transmembrane region in domain I. In addition to the genetic linkage analysis, we further characterized the truncated gene product by heterologous expression. No apparent function could be detected. Furthermore, we have also sought to answer why Na v 1.7 is essential for survival in rodents but not in humans. We find differences in the expression pattern of Na v 1.7 in central endocrine brain regions as well as the adrenal medulla. Our results add to the previous findings on congenital inability to experience pain ( 9 , 10 ) and further establish Na v 1.7 as a critical element of peripheral nociception in humans.
RESULTS
Pedigree and clinical features
The patients in the Canadian family presented with an absent response to pain. These patients did not report pain caused by infections or injury such as fractures and burns and were prone to joint, limb, oral cavity and nose damage. In their childhood, they never cried after an injury and the parents would only notice that the child was ill when they presented with a fever or obvious wounds. Their ability to discriminate between sharp and dull or hot and cold was intact. The neurological examination of the motor, sensory and mental function was normal. In particular, the patients were sensitive to light touch, coin distinction, pin prick, tickling, vibration, position and temperature. In addition, the patients had normal sweating. Three of the initially affected individuals (X1226, X1230 and X1377) claimed to start noticing pain in their early teens (12–14 years old), whereas the fourth individual (X1387), examined at age three, did not display pain behavior. The carriers did not present with the inability to experience pain phenotype.
Linkage analysis
Whole genome scan (WGS) two-point linkage analysis identified two loci with an LOD score > 2, namely, 3.74 at theta 0 for marker D6S1027 and 2.63 at theta 0 for marker D2S1776. Additional markers spanning 16.26 cM, flanking D2S1776, and 14.07 cM, proximal to D6S1027, were analyzed of both candidate loci. Lack of a consistent segregating homozygous haplotype excluded the chromosome 6 locus. Interestingly, further fine mapping on chromosome 2 resulted in an LOD score of 3.46 at theta 0 for marker D2S111, and haplotype analysis revealed that this marker is located in a large homozygous block (Supplementary Material).
The shared homozygous haplotype lies between markers D2S156 and D2S1267 (Supplementary Material), which span ∼13.7 Mb according to the May 2004 version of the genome browser, NCBI Build 35 ( http://genome.ucsc.edu/ ). The haplotype transmission pattern of the markers on chromosome 2 suggested that a single disease haplotype, presumably related to a single common ancestor, segregated with the affected individuals, all of whom are homozygous for the haplotype.
The identified region contains approximately 100 genes as defined by ENSEMBL. From these genes, bioinformatic and literature analysis further narrowed the potential number of prime candidates to 11 genes, including five voltage-gated sodium channels located in a cluster in the region. Because voltage-gated sodium channels have previously been found to be involved in altered nociceptive responses ( 1 ), SCN1A, SCN2A2, SCN3A, SCN7A and SCN9A were prioritized for sequence analysis.
Bidirectional sequencing of the coding regions was performed and a homozygous point mutation C984A (position relative to the ATG in reference sequence NM_002977) identified in exon 8 of SCN9A (Fig. 1B) . This point mutation results in the exchange of the amino acid Y328 for a stop codon, thus generating a truncation of the Na v 1.7 protein. Inheritance of the mutation across the pedigrees was consistent with it being causative to the phenotype (Fig. 1C) . This mutation was not present in DNA samples from 150 Caucasians, 48 Hispanics and 29 Asians.
Family tree and genetic linkage analysis. ( A ) The pedigree shows the degree of consanguinity in the Canadian family we have studied. Blackened symbols denote individuals with congenital inability to experience pain. The mode of inheritance is autosomal recessive with four affected males in the last generation, and clinical features as described in the text. ( B ) Sequence data showing the wild-type C allele in the pool of 25 Caucasian DNAs from healthy controls and the mutant A allele in an affected individual, as well as the sequence information for an asymptomatic carrier of the mutation. The arrows highlight the mutation from a C to an A. ( C ) Pedigree structure and genotype of Canadian family members with congenital inability to experience pain. Individuals with the congenital inability to experience pain phenotype are highlighted. ( D ) Schematic representation of the Na v 1.7 protein, and localization of the identified stop codon mutation after the fifth transmembrane region in domain I (arrow). The resulting product will lack all pore-forming regions of the channel.
Functional evaluation of the mutation
As indicated in Figure 1 D, the mutation will truncate Na v 1.7 shortly after the fifth transmembrane region of the channel. This is expected to result in either nonsense-messenger-mediated RNA decay, which is likely to have no additional functional consequences, or expression of a truncated protein lacking all pore-forming regions of the full-length channel.
To test for the eventual functionality of this truncated product, we re-engineered the mutation (Na v 1.7 stop328 ) and performed whole-cell patch-clamp experiments on transiently transfected HEK293 cells. As expected, transient transfection of the full-length human Na v 1.7 cDNA resulted in voltage-activated responses resembling previously published data ( 9 , 11 ) for heterologously expressed Na v 1.7 in all recordings from pZsGreen1-N1 fluorescent cells ( n = 18; Fig. 2 A). In contrast, transient transfection of Na v 1.7 stop328 did not result in measurable responses in any of the nine recorded cells, despite successful transfection of the studied cells as indicated by pZsGreen1-N1 fluorescence (Fig. 2 A).
Functional effects of mutation. ( A ) Transfection of HEK293 cells with Na v 1.7 stop328 did not produce currents despite successful transfection as indicated by pZsGreen1-N1 fluorescence (bottom). In contrast, HEK293 cells transfected with full-length Na v 1.7 showed typical voltage-activated sodium currents (top). ( B ) Transfection of Na v 1.7 stop328 into ND7/23 cells did not alter the function of the endogenous voltage-gated sodium current present in this cell line. For the examples shown, the cell transfected with Na v 1.7 stop328 had a capacitance of 32.3 pF, whereas the control cell, transfected with pZsGreen1-N1 only, had a capacitance of 33.7 pF. ( C ) Comparison of the I–V relationship for the endogenous voltage-gated sodium current in ND7/23. There was no effect of transfection of Na v 1.7 stop328 . ( D ) Transfection of Na v 1.7 stop328 had no effect on current density of the endogenous voltage-gated sodium current in ND7/23 cells.
Whereas these results indicate that the truncated protein is not functional as an ion channel by itself, the possibility exists that in nociceptive neurons, it would cause compensatory effects on expression of other voltage-gated sodium channels. To test for this possibility, we performed experiments on ND7/23 cells, a dorsal root ganglia (DRG)-derived neuroblastoma cell line that has been shown to endogenously express voltage-gated sodium currents ( 12 ). We transiently transfected ND7/23 cells, either with a mixture of Na v 1.7 stop328 and pZsGreen1-N1 or with pZsGreen1-N1 only.
As shown in Figure 2 B–D, pZsGreen1-N1-only transfected cells exhibited a large TTX-sensitive voltage-gated sodium current with a peak amplitude of 3472 ± 520 pA when stepped to − 15 mV from a pre-step holding potential of − 100 mV ( n = 13). When related to the surface area of the recorded cells, this corresponded to a current density of 147 ± 26 pA/pF. The voltage-dependence of activation in pZsGreen1-N1-only transfected cells was half-maximal at − 34 ± 3 mV, whereas the voltage-dependent inactivation in these cells was half-maximal at − 71 ± 2 mV. In ND7/23 cells transfected with hNa v 1.7 stop328 and pZsGreen1-N1, TTX-sensitive voltage-gated sodium currents that had properties indistinguishable to those recorded from pZsGreen1-N1-only transfected cells could be recorded. Thus, the recorded peak amplitude and current density were 3380 ± 451 pA and 152 ± 25 pA/pF, respectively ( n = 15). Likewise, half-maximal voltage-dependence of activation was at − 34 ± 4 mV and half-maximal voltage-dependence of inactivation was at − 72 ± 2 mV.
Comparison of Na v 1.7 expression between rodents and primates
Although the affected individuals in the Canadian family have a global deficiency of Na v 1.7, they appear otherwise healthy. In contrast, it has been described that a global deficiency of Na v 1.7 is lethal in mice shortly after birth ( 3 ), apparently caused by a failure to feed. Although it is possible that this difference is due to extensive parental care in humans, other factors could also be responsible. One possible explanation could be that Na v 1.7 serves slightly different functions in the two species. To test whether this could be the case, we compared the expression pattern of Na v 1.7 in rat, mouse, monkey ( Macaca fascicularis ) and human tissue using the in situ hybridization technique.
As expected, we detected strong hybridization signals for Na v 1.7 mRNA in all four species in neuronal cell bodies from different ganglia, including the dorsal root ganglia (Supplementary Material, Fig. S1). Likewise, we also detected light hybridization signal in all species in ventral horn spinal cord motor neurons (not shown). This confirmed that our probes were valid and that post-mortem human material is sufficiently viable for this type of analysis.
We next examined the expression of Na v 1.7 mRNA in brain and peripheral tissues from all four species. In both rodent species, we could detect a prominent mRNA hybridization signal in the paraventricular hypothalamic nucleus (Pa) and the supraoptic nucleus (SO) (Fig. 3 A and B). In contrast, such prominent labeling was not present in monkey or human brain in these regions (Fig. 3 C and D, respectively). In contrast, we were able to detect mRNA hybridization signal for beta-subunits of voltage-gated sodium channels in adjacent section of human brain tissue (not shown), indicating viability of the tissue for such studies. Furthermore, examination of autoradiograms from coronal and sagittal monkey brain sections failed to reveal any region with significant mRNA expression (data not shown). Weak labeling of a few individual cells in regions such as the Pa (Fig. 3 G and H) and SO (not shown) was detected in both primate species upon examination of emulsion-coated sections. However, the ISH signal observed in primates was much weaker than the strong labeling present in both rodent species, as illustrated for the Pa in Figure 3 E and F.
Na V 1.7 mRNA expression in rodent and primate brains. On autoradiograms, prominent mRNA hybridization signal can be observed in the paraventricular hypothalamic nucleus (Pa) and supraoptic nucleus (SO) in coronal brain sections from mouse ( A ) and rat ( B ). In contrast, note the absence of prominent labeling in the same regions on autoradiograms from monkey ( C ) and human ( D ) brains. On emulsion-coated sections, neurons expressing Na V 1.7 mRNA can be detected in both rodent (mouse, E ; rat, F ) and primate (monkey, G ; human, H ) brains in areas such as the Pa (E–H), although the signal intensity is considerably weaker in primates when compared with rodents. In (E–H), a high magnification bright-field photomicrograph of the Pa is presented. ac, anterior commissure; Arc, arcuate; CeA, central amygdaloid nucleus; DM, dorsomedial hypothalamic nucleus; cc, corpus callosum; Cx, cortex; Cd, caudate; opt, optic tract; EntCx, entorhinal cortex; HPT, hypothalamus; Pu, putamen; PV, paraventricular thalamic nucleus; VMH, ventromedial hypothalamic nucleus. Scale bars: 1 mm (A), 2 mm (B), 5 mm (C and D), 20 µm (E–H).
In addition, we detected mRNA expression in both the pituitary and adrenal glands in both rodent species, whereas these two glands were devoid of mRNA in humans (Fig. 4 ). It should be noted that in the same sections, expression was observed in the celiac ganglia (Fig. 4 B), confirming the presence of this ion channel within sympathetic ganglia in humans as well as the viability of the material.
Na V 1.7 mRNA expression in adrenal and pituitary glands. Na V 1.7 mRNA hybridization signal is detected in rat ( A ) but not in human ( B ) adrenal medulla; no labeling was observed over the adrenal cortex. Note that the celiac ganglionic cells (peripheral sympathetic ganglia; denoted by arrows) attached to the human adrenal gland express Na V 1.7 mRNA (B). Both the anterior and intermediate lobes of the rat pituitary gland ( C ) are labeled. In contrast, the anterior lobe of the human pituitary was devoid of Na V 1.7 mRNA hybridization signal ( D ). Scale bars: 1 mm (A and C), 2 mm (B and D).
These results indicate an apparent difference in the Na v 1.7 expression levels and pattern in rodents versus primates. This difference may explain why a global deficiency of Na v 1.7 is lethal in mice but not in humans.
DISCUSSION
Our study identifies and characterizes the molecular cause for congenital inability to experience pain in members of a Canadian family from Newfoundland. The identified truncation mutation in SCN9A did not produce voltage-gated sodium channel responses when transfected as a cDNA construct in HEK293 cells. Furthermore, transfection of the construct in a neuroblastoma cell line did not cause compensatory effects on other voltage-gated sodium channels. Differences in the expression pattern between rodents and primate species were observed, providing a possible explanation for why a global deficiency of Na v 1.7 is lethal in mice but not in humans.
The pedigree described here was also studied independently by another group who recently published their findings ( 10 ). Our results confirm their genetic linkage findings and further support that loss-of-function mutations in SCN9A are linked to a congenital inability to experience pain phenotype ( 9 , 10 ). In addition, our study adds some interesting aspects to these earlier studies.
Transfection of the truncation construct did not cause apparent compensatory effects on other voltage-gated sodium channels in ND7/23 cells. Thus, the truncated protein does not act as a dominant negative factor interfering with gating or current density produced by intact channels in this cell system. With regard to interpretations concerning the in vivo situation, it should be noted that ND7/23 cells, although DRG-derived, are not a native cell system. They thus could respond differently from native DRG neurons. Likewise, the relatively short duration of expression of proteins after a transient transfection may not be sufficient to cause major compensatory changes. Nevertheless, our findings also lend some experimental support for the idea that the observed clinical phenotype is a direct consequence of a global deficiency of Na v 1.7, rather than any secondary compensatory mechanisms.
Our study shows species differences in the expression pattern of Na v 1.7 that provide a possible explanation for why a global Na v 1.7 deficiency is lethal in rodents ( 3 ) but not in humans. It has been postulated that the lethality of the Na v 1.7 knockout mice is due to failure to feed, possibly resulting from dysfunctions of central, autonomic or enteric sensory neurons ( 3 ). Our finding of a distinct expression of Na v 1.7 in rodent brain and in endocrine glands (pituitary and adrenal) supports this hypothesis, because it suggests that Na v 1.7 contributes to the regulation of autonomic and endocrine function in rodent species. We have seen expression of Na v 1.7 in these rodent tissues not only by in situ hybridization, but also by immunohistochemistry, as described in a separate study (A. Morinville, B. Fundin, L. Meury, A. Jureus, K. Sandberg, J.J. Krupp, S. Ahmad and D. O'Donnell, submitted for publication). In contrast, the low expression levels of Na v 1.7 in the hypothalamus and apparent lack of expression in the adrenal and pituitary glands of primates, including humans, make it unlikely that in primates, Na v 1.7 plays a major role in the regulation of autonomic and endocrine function.
The observation that the patients in the Canadian family claim to start to feel pain during adolescence and thereon afterward may be reflective of the emergence of a true pain sensation during adolescence, but could also be a learned behavior. An emergence of pain sensation has not been described for the patients examined in the two other studies ( 9 , 10 ). Further clinical examinations would be desirable, but unfortunately we were not able to obtain consent from the subjects for such studies.
Our study adds to the long list of voltage-gated sodium channelopathies ( 13 ); for example, mutations in SCN1A and SCN2A cause different epilepsy syndromes ( 14–16 ), whereas mutations in SCN4A have been linked to myotonia conditions ( 17 ). Mutations in SCN5A are linked to Brugada syndrome ( 18 , 19 ), and recently, a truncation mutation in SCN8A that was associated with a case of cerebellar atrophy, ataxia, and mental retardation has been described ( 20 ).
Of note, although SCN3A, SCN9A, SCN10A and SCN11A have been implicated in different pain conditions ( 1–3 ), only SCN9A-linked channelopathies have been described so far, resulting in erythromelalgia ( 4 , 5 , 7 , 8 ), paroxysmal extreme pain disorder ( 6 , 15 ) and the lack of general pain sensation described by others ( 9 , 10 ) and, here, by us. This may indicate that Na v 1.7 plays a rather unique, yet essential, role in nociceptive processing. Such a function may be the transformation of passively conducted generator potentials into actively conducted action potentials in nociceptive nerve endings ( 3 , 7 , 8 ). Independent of whether Na v 1.7 fulfills this transformer function or not, our data further support the notion that Na v 1.7 is an extremely promising target for selective pharmaceutical intervention in the analgesia area.
MATERIALS AND METHODS
Patients, clinical evaluation and blood sample collection
The study family originates from Newfoundland, Canada. Their pedigree is shown in Figure 1 A, indicating an autosomal recessive mode of inheritance with four affected males in the last generation. Clinical evaluation of affected individuals was done by G.A.R.
DNA was obtained from family members displayed in Figure 1 , except for the parents in the first generation, i.e. 16 unaffected and four affected individuals. After obtaining informed consent, peripheral blood was collected from the family members for DNA extraction. In addition, lymphoblastoid cell lines were established for all patients ( 21 ). This study was approved by the local ethics board.
Genome scan analysis
Mutations in the nerve growth factor receptor gene TRKA have previously been causally linked to certain types of hereditary sensory and autonomic neuropathies ( 22 ). We thus first excluded the involvement of the TRKA gene by linkage analysis. Then a WGS with a 10 cM resolution consisting of 370 polymorphic markers was performed at the Genome Quebec Innovation Centre, Montréal. The data were analyzed with two-point parametric linkage analysis using the FASTLINK software program ( 23 ). The parameters for linkage analysis included a recessive mode of inheritance, a penetrance of 99%, a phenocopy rate of 1/100 000 and a disease frequency of 1.5/1000.
For further elucidation, 18 additional small tandem repeat (STR) markers were typed on chromosome 6, and seven additional STR markers on chromosome 2, around the peak markers.
Genotyping
Additional genotyping was performed by polymerase chain reaction (PCR) amplification of polymorphic markers at loci with an LOD score > 2 according to the Marshfield map ( http://research.marshfieldclinic.org/genetics/GeneticResearch/compMaps.as ).
Radiolabeled S 35 -dATP was incorporated into the product during the reaction, followed by electrophoresis of the product on 6% denaturing polyacrylamide gels. Bands were resolved on autoradiographic film. Haplotypes were constructed manually assuming minimal recombination.
Mutational analysis and bioinformatics
ENSEMBL genomic viewer ( www.ensembl.org ) was used for deriving a list of genes located between the flanking markers described earlier. This list was further reduced by literature analysis.
Subsequently, PCR primers were designed to amplify each exon and at least 30 bp of flanking sequence for nine candidate genes in the shared haplotype region, using primer 3 ( http://frodo.wi.mit.edu/primer3/primer3_code.html ). All PCR primers were tagged with M13 forward or M13 reverse sequence. Exons were amplified using AmplitaqGold polymerase under touchdown PCR cycling conditions ( 24 ): step 1: 95°C denaturing for 10 min; step 2: 94°C for 20 s, 61°C for 1 min (ramping –0.5°C per cycle), 72°C for 1 min (step 2 cycled 14 times); step 3: a further 20 cycles as above but without ramping at a fixed annealing temp of 54°C followed by a final elongation step of 72°C for 10 min. PCR products were sequenced in both directions by dye-terminator sequencing, using standard methods with M13 forward and reverse primers. Initial mutation detection was performed by comparing the sequence profiles of two affected individuals (X1387 and X1226) with a pool of 25 Caucasian DNAs from healthy controls. Alterations consistent with being causal were confirmed by segregation analysis, which included subsequent sequencing DNA from an additional affected individual (X1230), an unaffected sibling (X1231) and the parents of all siblings (X1225 and X1228, and X1229 and X1232). To further increase our confidence that the identified variant was not present in the general population, DNA samples from 150 Caucasians, 48 Hispanics and 29 Asians were also sequenced.
Constructs and site-directed mutagenesis
We re-engineered the identified stop codon into a cDNA clone of the human Na v 1.7 gene (amino acid sequence corresponding to NM_002977) assembled from overlapping PCR fragments. The mutant, Na v 1.7 stop328 , was generated by replacing the codon for residue 328 with a stop codon (QuickChange ® II Site-Directed Mutagenesis Kit, Stratagene). The resulting construct was cloned into a pcDNA™4/TO expression vector (Invitrogen). A pZsGreen1-N1 plasmid (Clontech) was used in co-transfections to visualize transfected cells.
Cell culture and electrophysiology studies
HEK293 cells were cultured in DMEM F-12 (Gibco) supplemented with 10% FBS at 37°C (5% CO 2 ), and split twice weekly using Trypsin. ND7/23 cells were cultured in DMEM + Glutamax (Gibco) supplemented with 10% FBS at 37°C (5% CO 2 ), and split twice weekly using Accutase. Both cell lines were plated 20 h before transfection into 25 cm 2 flasks, and transfected using Lipofectamine 2000 (Invitrogen AB) according to the manufacturer's description. Transfection was done either with a cDNA mixture (10:1 relation) of Na v 1.7 stop328 and pZsGreen1-N1 or with pZsGreen1-N1 only.
For electrophysiology experiments, cells were plated onto NUNC 35 mm culture dishes 20 h after transfection. Cells were allowed to recover from plating for an additional 20–48 h prior to experiments.
Electrophysiology studies
For whole-cell voltage clamp recordings, cells were continuously perfused at room temperature (22–24°C) with extracellular solution of composition (m m ): NaCl 137, KCl 5.0, CaCl 2 1.0, MgCl 2 1.2, HEPES 10, glucose 10, pH 7.4 (NaOH). Patch pipettes (Borosilicate glass; GC150, Harvard Apparatus Ltd, Edenbridge, Kent, UK) had resistances of 2–5 MΩ when filled with intracellular solution (m m ): Cs-gluconate 140, MgCl 2 1.2, HEPES 10, EGTA 10, pH 7.2 (CsOH). The experimental set-up was mounted on an inverted Nikon microscope with a 20-fold objective. Data were acquired using HEKA Pulse software in combination with a HEKA EPC10/2 amplifier (HEKA Electronic), or pClamp 9 software in combination with an Axopatch 200A amplifier using a Digidata 1200 interface (Molecular Devices, Union City, CA, USA). Currents were filtered at 10 kHz and digitized at 20 kHz. Series resistance was routinely compensated to 50%. Cell input resistance and membrane capacitance were continuously monitored using a short − 10 mV pulse. Standard holding potential for both HEK293 and ND7/23 cell experiments was − 100 mV. The voltage protocol consisted of 5 mV steps from − 90 to 85 mV. Analysis was done using HEKA PulseFit software or Clampfit software. All data are expressed as mean ± SEM.
Tissue collection and processing
All animals used in this study were treated according to procedures approved by the local AstraZeneca Animal Care Committee and in accordance with The Care and Use of Experimental Animals of Canadian Council on Animal Care.
Adult male Sprague–Dawley rats and C57/BL6 mice (Charles River, St Constant, Quebec, Canada) were sacrificed by decapitation. Brains, spinal cords with DRG attached, trigeminal ganglia (TGG) and peripheral tissues were rapidly dissected, snap-frozen at − 40°C in isopentane for 20 s and stored at − 80°C until sectioning. Macaca fascicularis ( n = 3) were obtained from ITR Laboratories Canada (Montreal, Canada). Male monkeys were pre-anesthetized with ketamine and euthanized with sodium pentobarbital; tissues were removed and processed as with rodent tissues. Frozen human tissues of brain, spinal cords, DRG, TGG and peripheral tissues (male, n = 7, postmortem delays of 4–17 h) were acquired from the Brain and Tissue Bank for Developmental Disorders (University of Maryland, Baltimore, MD, USA). All frozen tissues were transversely sectioned at 14 µm on a cryostat, and sections were thaw-mounted onto Superfrost Plus Slides (VWR, Montreal, Quebec, Canada) or Colormark Plus Microscope slides (Erie Scientific Company, Portsmouth, NH, USA). Slides were stored at − 80°C until further use.
Na v 1.7 probes
For the mouse probe, a 334 bp fragment corresponding to the early 3′ untranslated region (3′-UTR) of the SNC9A gene was amplified from mouse genomic DNA using 5′-GAAGGTGACTCACTCGTG-3′ as a forward primer and 5′-CATGTGCGCCTGAATTTC-3′ as a reverse primer. For the rat, a 397 bp fragment corresponding to the early 3′-UTR of the SNC9A gene was amplified from rat genomic DNA using 5′-TTGACAGCCTGTGAAGGTTG-3′ as the forward primer and 5′-TGAAATCGTAGCCAATCCTG-3′ as the reverse primer. For the human probe, a 302 bp fragment from the 3′-UTR was amplified from human genomic DNA using 5′-GTTTACAGCCTGTGAAAGTG-3′ as forward primer and 5′-GTGAAAAGATGACAAGGCAG-3′ as reverse primer. PCR amplification (reaction mix: 1 ng/µl of genomic DNA, 0.2 µ m of each dNTP, 1 m m of each primer, 0.1 U/µl of Taq DNA polymerase in 1 × Taq DNA polymerase buffer) was performed as follows: 3 min at 95°C, followed by 25 cycles of 30 s at 95°C, 45 s at 56°C and 30 s at 72°C. The amplicon was isolated on an agarose gel, using the QIAquick gel extraction kit (Qiagen, Mississauga, Ontario, Canada) and was ligated into pGEM-T-Easy vector (Promega, Nepean, Ontario, Canada). Plasmid from a single clone was purified using HiSpeed Plasmid midi kit (Qiagen).
The Na V 1.7 plasmids (mouse, rat, human) were linearized using appropriate restriction enzymes. The Na V 1.7 antisense probes (mouse, rat, human) were transcribed in vitro using either T7 or SP6 RNA polymerases (Promega) in the presence of 35 S-UTP and 35 S-CTP (Amersham Biosciences, Inc.). After transcription for 1 h at 37°C, the DNA template was digested using DNAse I (Amersham Biosciences, Inc.). Riboprobes were subsequently purified using ProbeQuant G-50 Micro Columns (Amersham Biosciences, Inc.). Probe quality was verified qualitatively by polyacrylamide-urea gel electrophoresis and scintillation counting.
In situ hybridization
In situ hybridization on selected tissues to detect Na V 1.7 mRNA was performed using a procedure previously described ( 25 ). Briefly, sections were thawed at room temperature prior to fixation for 10 min in 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4. After washing with 2 × standard sodium citrate buffer (2 × SSC), sections were first equilibrated with 0.1 M triethanolamine, pH 8.0, then treated with 0.25% acetic anhydride in triethanolamine, subsequently washed with 2 × SSC and finally dehydrated with increasing concentrations of ethanol. The 35 S-labeled Na V 1.7 probes (both antisense and sense) prepared as described earlier were diluted to an activity of 20 × 10 6 cpm/ml in a hybridization buffer containing 75% formamide, 600 m m NaCl, 10 m m Tris–HCl (pH 7.5), 1 m m EDTA, 1 × Denhardt's solution (Sigma, St Louis, MO, USA), 50 µg/ml denatured salmon sperm DNA (Sigma), 10% dextran sulfate (Sigma), 20 m m dithiothreitol. Slides were hybridized with the 35 S-labeled antisense probes for 18 h at 55°C in humidified chambers. Mouse and rat tissues were incubated with the mouse and rat antisense probes, respectively, whereas primate tissues (both monkey and human) were hybridized with the human probe. Both coronal and sagittal monkey brain sections were included in the analysis. Sections were washed with 2 × SSC and subsequently treated with 20 µg/ml RNase IA (Amersham Biosciences, Inc.) in RNase buffer (25 m m NaCl, 5 m m Tris–HCl, pH 7.5, 0.5 m m EDTA), pH 7.5, at 37°C for 45 min. Sections were then washed in a series of citrate buffers of increasing stringency (2 × SSC for 5 min at 37°C, 0.5 × SSC for 45 min at 65°C and finally 0.1 × SSC for 45 min at 70°C). After dehydration and drying, slides were exposed to Kodak Biomax MR film for 1–2 weeks. Slides were subsequently dipped in Kodak NTB2 emulsion diluted to 1:1 with distilled water and exposed for 8 weeks at 4°C before being developed with Kodak D-19 developer. Slides were then counterstained with hematoxylin and eosin.
Image analysis and photography
Hybridization signal intensity was assessed using both the film autoradiograms and the emulsion-dipped slides. The MCID image analysis system (Imaging Research, Inc., St Catharines, Ontario, Canada) equipped with a high resolution Xillix Microimager digital camera (Xillix Technologies Corp., Richmond, British Columbia, Canada) and a Nikon 55 mm camera lens was used to capture images of the film autoradiograms. The same settings were used for the acquisition of all images. Digital images of the autoradiograms were imported into Adobe Photoshop version 7.0 imaging software (Adobe Systems Inc., San Jose, CA, USA) for cropping and labeling. Brightfield and darkfield images of the emulsion-dipped slides were acquired using a Leica DMRB microscope (Leica Microsystems Inc., Richmond Hill, Ontario, Canada) equipped with Leica digital camera 300 and image acquisition software. Digital images were imported into Adobe Photoshop (Adobe Systems Inc.) for cropping, labeling and contrast adjustments.
SUPPLEMENTARY MATERIAL
Supplementary Material is available at HMG Online.
ACKNOWLEDGEMENTS
We wish to thank Dr E. Ives at Memorial University, St John's, Newfoundland, Canada, for her aid in this study. We also would like to thank the members of the Canadian family studied, for their cooperation. Support by the AstraZeneca DNA sequencing service at Alderley Park is gratefully acknowledged. This study was funded by AstraZeneca.
Conflict of Interest statement . S.A., L.D., A.B.E., D.H., U.K., P.-E.L., L.M., T.M., A.M., A.M., J.M., D.O'D., C.R., H.S. and J.J.K. are employees of AstraZeneca.
