Introduction.
Odontogenic pain is usually caused by either noxious physical stimuli or by the release of inflammatory mediators that stimulate receptors located on terminal endings of nociceptive afferent nerve fibres (Hargreaves & Dubner 1992). Nociceptive fibres are distributed throughout the body (Kumazawa 1996) and are particularly prevalent in the trigeminal nerve which innervates dental pulps and periapical tissues (Hargreaves 1998). However, the exact role of chemical mediators in the pathogenesis of toothache remains unclear.
The diagnosis of pulpal pain can be difficult, because the site of the pain often cannot be precisely located (Dummer et al . 1980). This is due, in part, to the fact that there are two types of afferent nerve fibres within the pulp which are activated in different ways (Närhi et al . 1996). Myelinated A-fibres, stimulated by cold, heat or drilling cause fast, well localized pain, whereas unmyelinated C-fibres, activated by stimuli which cause damage to pulp tissue produce dull, poorly localized pain (Närhi et al . 1992). It is now known that these C-fibres respond to chemical mediators released as a result of tissue damage (Ngassapa 1996).
Recently, the subgroup of sensory nerves containing substance P (SP), neurokinin A (NKA) and calcitonin gene-related peptide (CGRP) has been the subject of much research (Silverman & Kruger 1987, Olgart et al. 1993, Kerezoudis et al . 1994, Buck et al . 1999). SP, NKA and CGRP are abundant in the pulp and periodontium (Kim 1990, Ohkubo et al . 1993) and are the only pulpal peptides to date which have been shown to originate from the trigeminal ganglion (Gazelius et al . 1987, Wakisaka 1990). The peptides contained in these neurones are found in unmyelinated C-fibres and probably some Adelta fibres. On excitation, these nerves release neuropeptides, producing an inflammatory response which is often called neurogenic inflammation (Györfi et al . 1993). SP, NKA and CGRP released by sensory fibres exert profound effects on blood flow, inflammatory and immune responses, and connective tissue cells and thus may be significant in many inflammatory conditions including pulpal pain (Gazelius et al . 1987, Wakisaka 1990, Olgart 1996, Pashley 1996).
There is good evidence to show that peptides are important in neurogenic inflammation but their role in pulpal pain remains unclear. Therefore, the aim of this study was to compare the levels of SP, NKA and CGRP in human dental pulp tissue from painful and non-painful teeth.

Materials and methods.

Patient selection.
The study was approved by the local Research Ethics Committee of Queen’s University Belfast and all subjects ( n = 66) gave their written informed consent. The pain group consisted of 46 subjects (28 male and 18 female patients; mean age 32.3, range 18–71 years) complaining of toothache, who attended a dental hospital for treatment of their dental pain. The control group comprised 20 subjects (10 male and 10 female patients; mean age 13.9, range 11–16 years), who required extraction of healthy teeth for orthodontic reasons.

Inclusion and exclusion criteria.
The inclusion criteria for the pain group were that the subjects were adults (18 years or older), diagnosed as having acute irreversible pulpitis with no evidence of periapical inflammation and no previous treatment for the symptomatic tooth.
Exclusion criteria were conditions requiring antibiotic prophylaxis or additional treatment procedures, debilitating disease, chronic pain, diabetes, haematological disorders or a history of taking centrally acting drugs (e.g. tricyclic antidepressants), known to interfere with the release of neuropeptides, within the previous 6 months.

History and examination.
Demographic details, the history of the pain, the medical history (including full details of any analgesic medication) and the patient’s smoking habits were recorded. Routine clinical and radiographic examinations were performed prior to treatment. Pain levels were assessed using a visual analogue scale (VAS). Each patient quantified the intensity of pain by placing a mark on a 10-cm VAS running from ‘no pain’ to ‘unbearable pain’.

Collection of pulp tissue from endodontically treated painful teeth.
Local anaesthesia (2% lignocaine with 1 : 80 000 adrenaline) was administered and an access cavity was prepared under rubber dam isolation. Pulp tissue was extirpated using a barbed broach or Hedström-file, placed into a preweighed Eppendorf tube, immediately frozen in liquid nitrogen and stored at –70 C (Gazelius et al . 1987). The root canals were thoroughly irrigated with 1% sodium hypochlorite, dried with paper points and dressed with either calcium hydroxide ( n = 15) or Ledermix paste (Lederle Laboratories, Gosport, UK) ( n = 25). The access cavity was then sealed with a cotton pellet and glass ionomer cement until the next visit.

Collection of pulp tissue from painful and non-painful extracted teeth.
Six patients choose to have extraction of the painful teeth instead of root canal treatment. All extractions of painful and non-painful teeth were performed under local anaesthesia (2% lignocaine with 1 : 80 000 adrenaline) with no complications. All extractions were completed in less than 2.5 min. Immediately after extraction, each tooth was split in a vice fitted with a cutting edge and the pulp tissue was removed using a barbed broach or Hedströmfile within 1–2 min. The pulp was then placed in a preweighed Eppendorf tube, immediately frozen in liquid nitrogen and stored at –70 C.

Extraction of neuropeptides from pulp tissue.
The extraction methodology was based on the results obtained from a previous study (Awawdeh et al . 1999). Briefly, the Eppendorf tube was weighed and the pulp weight was calculated. The required amount of 0.5 mol L 1 acetic acid was measured in a ratio of 8 mLg –1 tissue and brought to boiling in a water bath. The pulp tissue, whilst still in a partly frozen condition, was immersed in the acetic acid for 10 min. The specimen was then centrifuged (2200 g , 20 min, 4 C) and the supernatant was collected and stored at –70 C.


Table 1. Concentrations of SP-Ir, NKA-Ir and CGRP-Ir in painful and non-painful pulp tissue.

Radioimmunoassay (RIA).
The substance P-immunoreactivity (SP-Ir), neurokinin A-immunoreactivity (NKA-Ir), and calcitonin gene-related peptide-immunoreactivity (CGRP-Ir) were determined by radioimmunoassay (RIA) using antisera SP 152(2), NKA 570(3) (Wellcome Research Laboratories, Queen’s University, Belfast, UK) and CGRP RAS6012 (Peninsula Laboratories, St. Helens, UK). All radioactive tracers were supplied by Amersham International (Little Chalfont, UK).
The assay system consisted of a total volume of 300 L, comprising 100 L antiserum, 100 L standard (9.5–625 pg mL –1 ) or 100 L pulp tissue extracts and 100 L of radioactive tracer [I 125 ].
The sensitivity and cross-reactivity of antisera SP 152 (2) and NKA 570 (3) have previously been reported (Maule et al . 1989). The sensitivity of antisera CGRP RAS6012 was 0.1 pg mL –1 and the cross reactivity has been reported previously (Tools for Protein and Peptide Research, Product Catalogue, Peninsula Laboratories, St. Helens, UK).
The SP-Ir assay used sodium phosphate buffer (0.04 mol L 1 , pH 7.4), containing 0.02% (w/v) thiomersal, 0.14 mol L 1 sodium chloride and 0.2% (w/v) bovine serum albumin (radioimmunoassay grade) in all dilutions of samples, antiserum, radioactive label and standards. Standards and extracted pulp tissue samples were incubated with antiserum 152(2) (final dilution of 1 : 600 000) for 24 h at 4 C. The I 125 -SP radioactive tracer (IM175, Amersham International, labelled using Bolton/Hunter reagent) was added, and the specimens incubated for a further 24 h at 4 C.
The NKA-Ir assay protocol was similar to that for SP-Ir, except that antiserum NKA 570(3) was used at a final dilution of 1 : 126 000 and the specimens incubated with radioactive tracer for 48 h.
The CGRP-Ir assay used sodium phosphate buffer (0.4 mol L 1 , pH 7.4) containing 0.02% (w/v) thiomersal, 0.05 mol L 1 sodium chloride, 0.1% (w/v) bovine serum albumin, and 0.1% (v/v) triton in all dilutions of the pulp tissue extracts, antiserum, radioactive label and standards. The antiserum RAS6012 was used at a final dilution of 1 : 900 000 and the specimens were incubated with radioactive tracer for 48 h.
A similar separation procedure was used for all three neuropeptides. The bound and free counts were separated by the addition of 1 mL of 0.5% (w/v) dextran-coated charcoal pellets in 0.4 mol L 1 sodium phosphate buffer, pH 7.4, containing 10% (v/v) horse serum (Gibco, Grand Island Biological Co, NY, USA). The specimens were then centrifuged (2000 g , 25 min, 4 C) and the charcoal pellets counted on a Cobra II Gamma Counting System (Packard Acanberra Company, Meriden, CT, USA).
All samples were assayed in duplicate and the mean values calculated. Individual controls for apparent binding in the absence of antibody were included in all assays. Radioimmunoassay results were expressed as nanogram of immunoreactive peptide per gram (ng g 1 ) of wet tissue.

Statistical analysis.
The concentrations of SP-Ir, NKA-Ir and CGRP-Ir were highly variable (Table 1) and a square root transformation was performed to produce normal distributions prior to analysis using parametric tests. The level of statistical significance was set at P < 0.05. Independent-Student’s t -tests were used to analyse differences between the neuropeptide levels in painful and non-painful teeth. Pearson’s product moment correlation coefficients were calculated for the three neuropeptides studied.
Stepwise regression analysis was used to determine whether age, gender, smoking, analgesic medication prior to treatment, tooth type (anterior, posterior), tooth location (maxillary, mandibular) or the presence of pulpal pain were independent factors affecting the levels of SP-Ir, NKA-Ir and CGRP-Ir.

All patients in the pain group reported a VAS score of 5 or more at their initial appointment and then reported a score of 0 on their second visit, indicating total relief of pain following pulpectomy regardless of the intracanal treatment. This also confirmed the diagnosis of pulpal pain.

Neuropeptide levels in pulp tissue samples.
Both SP-Ir and CGRP-Ir were present in all samples with NKA-Ir in 96% of the 66 pulp samples. The mean concentrations of SP-Ir, NKA-Ir and CGRP-Ir were significantly higher in pulp tissue from painful teeth compared with non-painful teeth (Table 1, Fig. 1). Notably, the concentration of NKA was more than four times higher in painful teeth than in non-painful teeth. The mean concentration of CGRP was much higher than both SP-Ir and NKA-Ir in both painful and non-painful teeth.


Figure 1. Neuropeptide concentration (ngg–1) in pulp tissue from painful and non-painful teeth.

There were moderate but statistically significant correlations between the levels of the three neuropeptides in both painful and non-painful teeth. In painful teeth, Pearson’s product moment correlation coefficient ( r ) for SP-Ir/NKA-Ir was r = 0.60 ( P < 0.0001), SP-Ir/CGRP-Ir was r = 0.54 ( P < 0.0001) and NKA-Ir/CGRP-Ir was r = 0.42 ( P = 0.004). The correlations were stronger in non-painful teeth; SP-Ir/NKA-Ir was r = 0.59 ( P = 0.006), SP-Ir/CGRP-Ir was r = 0.72 ( P < 0.0001) and NKA-Ir/CGRP-Ir was r = 0.61 ( P = 0.004).
The effect of age, gender, tooth location, tooth type (incisor, canine, premolar and molar) and smoking on levels of SP-Ir, NKA-Ir and CGRP-Ir in the pulp tissue was studied. Smoking was the only variable, apart from pain, which emerged from the stepwise regression analysis as having a relationship with neuropeptide concentration. Bivariate analysis indicated that the levels of CGRP-Ir, but not SP-Ir or NKA-Ir, were significantly ( P = 0.02) higher in smokers compared with non-smokers (Table 2).


Table 2. Mean (SD) concentrations of SP-Ir, NKA-Ir and CGRP-Ir in painful teeth from smokers and non-smokers.

Regression analysis showed a significant association between the severity of pain recorded using the VAS and the levels of SP-Ir ( P = 0.036) and NKA-Ir ( P < 0.001) but not CGRP-Ir ( P = 0.313).

Discussion.
This study is the first to compare the concentrations of SP, NKA and CGRP in pulp tissue from painful and non-painful human teeth. All three neuropeptides were identified in virtually all of the pulp tissue samples, whether from painful or non-painful teeth. The results of the present study showed that the levels of all three neuropeptides were significantly higher in pulp samples from painful teeth compared with healthy control teeth. The levels, in descending order, were CGRP > NKA > SP. The increased levels of these neuropeptides strongly suggest that afferent peptidergic nerve fibres in the dental pulp do not simply conduct impulses centrally, but also play an active and important part in the overall response of the pulp to injury.
The presence of the neuropeptides in non-painful teeth suggests nonsensory functions. In support of this, SP, NKA and CGRP have been reported to be involved in the regulation of pulpal blood flow in cats and ferrets (Olgart et al . 1993, Berggreen & Heyeraas 1999). In healthy tissues, neurovascular mediators are present in small amounts to maintain vascular tone, ensure smooth flow of blood and the consistent supply of nutrients to the tissue, and to regulate the interstitial pulpal pressure within a noncompliant tissue (Goodis & Saeki 1997, Fristad 1997).
Neuropeptides are involved in neurogenic inflammation (Alstergren et al . 1995) and their increased levels in pulpitis may be an attempt to regulate pulpal blood supply and thus control the fluid exudate associated with the inflammatory process. The release of peptides such as SP, NKA and CGRP has long been implicated in neurogenic inflammation and many studies support the hypothesis that neuropeptides are involved in the pathophysiology of inflammatory diseases (Maggi 1995, Alstergren et al . 1995, Olgart 1996). Most previous studies on neuropeptide responses to pulpal injury have been carried out in animals and have demonstrated that neuropeptides are altered in inflamed pulps (Byers 1994, Buck et al . 1999). The relevance of neuropeptides to humans is less clear and caution must be exercised when extending the results of animal studies to humans. Nevertheless, the fact that SP, NKA and CGRP, which are found in many species, are highly conserved molecules in evolution suggests that they may play very similar roles in a wide variety of species (Silverman & Kruger 1987).
Although animal models of pulpal injury are fundamental to our current understanding of neuropeptides in pulpal inflammation (Byers 1994, Buck et al . 1999), it is important to extend such research to examine the role of neuropeptides in human pulpitis. Since ethical considerations are of paramount importance in research involving human subjects, it is not always possible to achieve closely matched patient groups. In the current study the control group was recruited from young patients requiring extractions for orthodontic reasons, whereas the pain group patients, who attended a casualty clinic for extraction or the first stage of root canal treatment, had a much greater age range. Although third molars were considered as an alternative source of healthy pulp tissue from age-matched individuals, this source was rejected because of difficulties associated with the extraction of these teeth coupled with the fact that teeth which are not in occlusion are known to have lower neuropeptide levels than normal (Byers & Närhi 1999). Furthermore, since logistic regression analysis of the data from the current study indicated that age was not significantly related to neuropeptide concentration, the age difference in the subject groups was considered not to have introduced a discernible biological effect. It is also acknowledged that by virtue of patient treatment, the method of harvesting pulp tissue was different for painful teeth compared with non-painful ones. Most of the painful pulp tissue samples were extirpated during root canal treatment, with the exception of six samples that were obtained from extracted painful teeth. In non-painful teeth the pulp was removed after the tooth had been extracted and split. This difference is unlikely to have affected the neuropeptides studied, because both procedures took the same amount of time (~3 min). Furthermore, any neuropeptides discharged from nerves during either tooth extraction or pulpectomy would have accumulated in the pulp tissue which was collected and boiled as a whole.
There are no published reports of quantitative studies on the levels of SP, NKA and CGRP in pulp tissue from painful human teeth to compare with the present investigation. Only four studies have reported the levels of SP, NKA and CGRP in non-painful human teeth (Brodin et al . 1981, Parris et al . 1989, Goodis & Saeki 1997, Pertl et al . 1997). In the present study, the levels of SP and NKA in healthy pulps were comparable but CGRP was substantially more abundant. This agrees with previous studies that reported similar levels of SP and NKA (Goodis & Saeki 1997) and a 10-fold increase in the pulpal concentration of CGRP over SP (Pertl et al . 1997). The higher levels of CGRP compared with SP and NKA in healthy pulps can probably be explained by a greater number of nerve fibres containing CGRP and it has been shown that trigeminal CGRP neurones are more numerous than SP neurones in animals (Dalsgaard 1988) and in humans (Matsuyama et al . 1986).
There is evidence that SP (and possibly NKA) are responsible for the initiation of vasodilation, whilst CGRP mediates the late and more dominant phase of vasodilation in pulpal inflammation (Kerezoudis et al . 1994, Olgart 1996, Berggreen & Heyeraas 1999). In addition, CGRP has been shown to potentiate the actions of SP (Greves et al . 1985) and is believed to have an important role in pulpal repair following injury (Byers & Taylor 1993). For example, it is known that CGRP stimulates human endothelial cell proliferation (Hægerstrand et al . 1990). The inhibition of T-lymphocyte proliferation by CGRP suppresses the inflammatory response, whereas SP has an opposite stimulatory effect (Umeda et al . 1988). It is possible therefore that the high levels of CGRP in painful teeth may modify the inflammatory and immune processes and thus promote healing. The abundance of CGRP in both health and disease suggests that CGRP has a role in inflammation, healing and regeneration within the dental pulp.
The proportional increase in NKA levels in painful pulps compared with healthy pulps was greater than that of either SP or CGRP. Both NKA and SP evoke the release of pro-inflammatory cytokines from human monocytes but NKA is more potent in this regard ( Lotz et al . 1988). Because NKA coexists with SP it has been tentatively concluded that they have a similar role (Wakisaka 1990). The greater proportional increase in NKA suggests an important role for this neuropeptide in promoting pulpal inflammation. However, further investigations are required to elucidate the precise function of NKA in the dental pulp.
The concentration of CGRP (but not SP or NKA) in painful teeth was significantly higher in smokers compared with non-smokers. Nicotine triggers the release of SP, NKA and CGRP from C-fibres in the lungs (Hong et al . 1995) and a significant increase in NKA and CGRP has been reported in the pulmonary effluent in guinea pigs in response to smoke inhalation (Lee et al . 1995). Therefore, it seems probable that nicotine causes release of CGRP from afferent nerves in human pulp tissue.

References.

Alstergren P, Appelgren A, Appelgren B, Kopp S, Lundeberg Y, Theodorsson E (1995) Co-variation of neuropeptide Y, calcitonin gene-related peptide, substance P and neurokinin A in joint fluid from patients with temporomandibular joint arthritis. Archives of Oral Biology 40, 127-35.
Awawdeh L, Lundy FT, Shaw C, Lamey P-J, Linden GJ, Kennedy JG (1999) A comparison of four extraction methods for substance P, neurokinin A and calcitonin gene-related peptide from human dental pulp tissue. Archives of Oral Biology 44, 999-1004.
Berggreen E, Heyeraas KJ (1999) The role of sensory neuropeptides and nitric oxide on pulpal blood flow and tissue pressure in the ferret. Journal of Dental Research 78, 1535-43.
Brodin E, Gazelius B, Olgart L, Nilsson G (1981) Tissue concentration and release of substance P-like immunoreactivity in the dental pulp. Acta Physiologica Scandinavica 111, 141-9.
Buck S, Reese K, Hargreaves KM (1999) Pulpal exposure alters neuropeptide levels in inflamed dental pulp and trigeminal ganglia: Evaluation of axonal transport. Journal of Endodontics 25, 718-21.
Byers MR (1994) Dynamic plasticity of dental sensory nerve structure and cytochemistry. Archives of Oral Biology 39 (Suppl. 1), 13-21.
Byers MR, Närhi M (1999) Dental injury models: Experimental tools for understanding neuroinflammatory interactions and polymodal nociceptor functions. Critical Reviews in Oral Biology and Medicine 10, 4-39.
Byers MR, Taylor PE (1993) Effect of sensory denervation on the response of rat molar pulp to exposure injury. Journal of Dental Research 72, 613-8.
Dummer PMH, Hicks R, Huws D (1980) Clinical signs and symptoms in pulp disease. International Endodontic Journal 13, 27-35.
Fristad I (1997) Dental innervation: functions and plasticity after peripheral injury. Acta Odontologica Scandinavica 55, 236-54.
Gazelius B, Edwall B, Olgart L, Lundberg JM, Hökfelt T, Fischer JA (1987) Vasodilatory effects and coexistence of calcitonin generelated
peptide (CGRP) and substance P in sensory nerves of cat dental pulp. Acta Physiologica Scandinavica 130, 33-40.
Goodis HE, Saeki K (1997) Identification of bradykinin, substance P, and neurokinin A in human dental pulps. Journal of Endodontics 23, 201-4.
Greves LP, Nyberg F, Ternius L, Hökfelt T (1985) Calcitonin generelated peptide is a potent inhibitor of substance P degradation. European Journal of Pharmacology 115, 309-11.
Györfi A, Fazekas A, Irmes F, Jakab G, Suto T, Rosivall L (1993) Role of substance P (SP) in the development of symptoms of neurogenic inflammation in the oral mucosa of the rat. Journal of Periodontal Research 28, 191-6.
Hægerstrand A, Dalsgaard CJ, Jonzon B, Larsson O, Nilsson J (1990) Calcitonin gene-related peptide stimulates proliferation of human endothelial cells. Proceedings of the National Academy of Science of the United States of America 87, 3299-303.
Hargreaves KM (1998) Endodontic pharmacology. In: Cohen S, Burns RC, eds. Pathways of the Pulp. St. Louis, USA: Mosby, 600-7.
Hargreaves KM, Dubner R (1992) Mechanisms of pain and analgesia. In: Dionne RA, Phero J, eds. Management of Pain and Anxiety in Dental Practice. New York: Elsevier Press, 18-40.
Hong J-L, Rodger IW, Lee L-Y (1995) Cigarette smoke-induced bronchoconstriction: cholinergic mechanisms, tachykinins, and cyclooxygenase products. Journal of Applied Physiology 78, 2260-6.
Kerezoudis NP, Olgart L, Edwall L (1994) CGRP (8-37) reduces the duration but not the maximal increase of antidromic vasodilation in dental pulp and lip of the rat. Acta Physiologica Scandinavica 151, 73-81.
Kim S (1990) Neurovascular interactions in the dental pulp in health and inflammation. Journal of Endodontics 16, 48-53.
Kumazawa T (1996) The polymodal receptor- bio-warning and defence system. Progress in Brain Research 113, 3-18.
Lee L-Y, Lou YP, Hong J-L, Lundberg JM (1995) Cigarette smokeinduced bronchoconstriction and release of tachykinins in guinea pig lungs. Respiratory Physiology 99, 173-81.
Lotz M, Vaughan JH, Carson DA (1988) Effect of neuropeptides on production of inflammatory cytokines by human monocytes. Science 241, 1218-21.
Maggi CA (1995) Tachykinins and calcitonin gene-related peptide (CGRP) as co-transmitters released from peripheral endings of sensory nerves. Progress in Neurobiology 45, 1-98.
Matsuyama T, Wanaka A, Yoneda S, Kimura K, Kamada T, Girgis S et al. (1986) Two distinct calcitonin gene-related peptide-containing peripheral nervous systems: distribution and quantitative difference between the iris and cerebral artery with specific reference to substance P. Brain Research 373, 205-12.
Maule A, Shaw C, Halton D, Johnston C, Fairweather I, Buchanan K (1989) Tachykinin immunoreactivity in the parasitic flatworm Diclidophora merlangi and its fish host the whiting (Merlangius merlangus): radioimmunoassay and chromatographic characterisation using region-specific substance P and neurokinin A antisera. Comparative Biochemistry and Physiology. A: Comparative Physiology 94, 533-41.
Närhi M, Jyväsjärvi E, Virtanen A, Huopaniemi T, Nagassapa D (1992) Role of intradental A and C type nerve fibres in dental pain mechanisms. Proceedings of the Finnish Dental Society 88 (Suppl. 1), 507-16.
Ngassapa DN (1996) Comparison of functional characteristics of intradental A- and C- nerve fibres in dental pain. East African Medical Journal 73, 207-9.
Ohkubo T, Shibata M, Yamada Y, Kaya H, Takahashi H (1993) Role of substance P in neurogenic inflammation in the rat incisor pulp and the lower lip. Archives of Oral Biology 38, 151-8.
Olgart LM, Edwall L, Fried K (1993) Cat dental pulp after denervation and subsequent reinnervation: changes in blood flow regulation and distribution of neuropeptide, GAP-43, and low affinity neurotrophin receptor-like immunoreactivity. Brain Research 625, 109-19.
Parris WG, Tanzer FS, Fridland GH, Harris EF, Killmar J, Desiderio DM (1989) Effects of orthodontic force on methionine enkephalin and substance P concentrations in human pulpal tissue. American Journal of Orthodontics and Dentofacial Orthopedics 95, 479-89.
Pashley DH (1996) Dynamics of the pulpo-dentine complex. Critical Reviews in Oral Biology and Medicine 7, 104-37.
Pertl C, Amann R, Odell E, Robinson PD, Kim S (1997) Effects of local anaesthesia on substance P and CGRP content of the human dental pulp. Journal of Endodontics 23, 416-8.
Silverman JD, Kruger L (1987) An interpretation of dental innervation based upon the pattern of calcitonin gene-related peptide (CGRP)-immunoreactive thin sensory axons. Somatosensory and Motor Research 5, 157-75.
Umeda Y, Takamiya M, Yoshizaki H, Arisawa M (1988) Inhibition of mitogen-stimulated T lymphocyte proliferation by calcitonin gene-related peptide. Biochemical and Biophysical Research Communications 154, 227-35.
Wakisaka S (1990) Neuropeptides in the dental pulp: Distribution, origins, and correlation. Journal of Endodontics 16, 67-9.