E. Z. DAJANI, 1, 2, T. G. SHAHWAN, 1, 3, N. E. DAJANI, 1, 3



PROSTAGLANDINS AND BRAIN-GUT AXIS


1 International Drug Development Consultants (IDDC) Corporation, Long Grove, Illinois.
2 Gastroenterology Section, Loyola University Medical Center, Maywood, Illinois.
3 University of Illinois School of Medicine, Peoria, Illinois, USA



  Prostaglandins (PGs) have well documented physiological and pharmacological actions on the gastrointestinal (GI) tract. This communication reviews the evidence for peripheral and central nervous system (CNS) physiological actions of PGs in order to determine their role in the brain-gut axis, if any. PGs are widely distributed in nearly all cells peripherally and centrally. Laboratory and clinical evidence indicate that there is a direct relationship between altered GI physiological functions and peripheral PGs biosynthesis. Either local or parenteral administration of natural E-series PGs alters GI physiological functions particularly those relating to mucosal defense. Furthermore, the cyclooxygenase enzymes (COX), which are responsible for the PGs biosynthesis, have been localized in the brain as well as peripherally. However, increased levels of PGs in the brain have been associated with pathological processes such as inflammation, pain, fever and addiction. Although PGs have been shown to modulate CNS effects of catecholaminergic, serotoninergic and cholinergic neurons, there is no meaningful information concerning their direct central effect on GI function. The evidence for a clear physiological role of central PGs on the GI tract is not convincing. At this time, we conclude that PGs primarily manifest their activity on the GI tract by peripheral rather than by central mechanisms.

Key words:    Addiction, CNS, COX, cyclooxygenase, cytoprotection, fever, inflammation, mucosal protection, pain, prostaglandins, PGE2, PGF2alpha, PGD2, thromboxane B2, TXB2, stress ulcers



INTRODUCTION

Prostaglandins constitute a family of unsaturated fatty acids with 20-carbon skeleton. Prostaglandins are found in almost every mammalian cell and are considered locally acting hormones (autacoids). They are synthesized on demand and subsequently inactivated at or near the sites of their synthesis. Prostaglandins are metabolically unstable compounds and are not stored. Because of their omnipresence in nearly all cells, PGs possess wide variety of physiological and pharmacological action, which are occasionally troublesome in their clinical use as drugs (1). There is clear evidence that PGs have local and systemic action affecting GI physiology (2). However, the role of central PGs in the regulations of GI physiology and particularly mucosal defense is not clearly understood. The purpose of this communication is to review the evidence for a CNS role of PGs in GI physiology and pathophysiology in order to examine their role on the brain-gut axis, if any.

Brain-Gut Axis Concept

The concept supporting the existence of functionally important “brain-gut axis” was originally proposed to account for the fact that several peptides including bombesin, neurotensin and calcitonin-gene-related peptide occur both in brain and gut and seem to exert opposite actions on gut function when administered centrally and peripherally (3 - 5). Support for this concept is derived from clinical and laboratory studies showing that stress ulcer formation could be prevented by anxiolytic and antidepressant drugs. The reduction of anxiety, which is associated with the stressful stimuli, occurs by a direct effect on the CNS as shown by the inhibitory effects of a centrally administered imipramine on stress ulcer formation (6, 7). Furthermore, the phenothiazine tranquilizer thiopropazoate had been demonstrated to significantly potentiate the anti-ulcer action of cimetidine (a histamine-H2-receptor antagonist) and propantheline (a peripheral anticholinergic drug) against stress ulcers (8). These laboratory studies indicate that the CNS mediated pharmacological actions of thiopropazoate potentiated the peripheral GI protective action of the anti-ulcer drugs cimetidine and propantheline, which provides further support for the concept of brain-gut axis (8).

Several lines of evidence indicate the involvement of dopamine in peripheral and central action of many drugs affecting the brain-gut axis (5, 9). As is shown in Table 1, there is laboratory and clinical evidence that peripheral and central dopamine deficiency is associated with duodenal ulcers (9, 10). Since prostaglandins and other eicosanoids are involved in the presynaptic release of the neurotransmitters dopamine and serotonin (11), it is possible that some of their physiological action on the gut may be mediated by these neurotransmitters. However, the effects of central administration of PGs on dopamine-mediated effects on the GI tract have not yet been adequately investigated.

Table.1. Prostaglandins, Central Dopamine, Peptic Ulcer and Brain-Gut Axis
Szabo S, et al. (9, 10).

Although PGs, growth factors and hormones possess direct cellular protective effects on the GI tract that is independent of a central influence, there is preliminary evidence which indicates that the CNS plays a contributory role towards this cytoprotection. In their cytoprotective study with gastric mucosal cells, Bodis et al. (12) showed that intact peripheral innervations are needed for the maximum demonstration of the prostacyclin-induced gastric cytoprotection.

E-Prostaglandins and Gastrointestinal Physiology

Prostaglandins of the E-series are the principal autacoids localized in the GI tract and have several well-characterized physiological actions (1). E-series prostaglandins inhibit basal and stimulated acid secretion and protect the GI mucosa from injury induced by noxious agents. E- and F-series PGs have opposing dose-related effects on the lower esophageal sphincter and circular intestinal muscle causing relaxation and contractions, respectively (1, 13, 14). Other physiological effects of PGEs include an increase in hepatic blood flow, contraction of the gallbladder, relaxation of the sphincter of oddi, inhibition of pancreatic secretion and insulin release, and reduced absorption and induced secretion of electrolytes and water in the jejunum, and ileum, but not the colon (Table 2).

Table.2. Selected Peripheral Gastrointestinal Physiological Effects of Gut PGs. Data were adapted from Dajani (1).
aThe dosages shown effective in animal cytoprotective studies are well below their gastric antisecretory doses.

A direct relationship exists between altered GI physiological function and prostaglandin synthesis (15, 16). For example, the nonsteroidal anti-inflammatory drugs (NSAIDs)-induced PGs depletion in the GI tract results in gastroduodenal ulceration and/or ulcer related GI complication (Table 3; 17). The administration of natural or synthetic PGEs, either by parenteral, oral or local routes, can overcome the GI toxicity associated with the mucosal depletion of PGE’s induced by NSAIDs (18 - 20). In addition, systemic or topical administration of natural and synthetic PGEs analogs can reproduce their well-characterized GI physiological actions on the inhibition of acid secretion and mucosal protection (2, 20). The fact that mucosal protection by PGs is demonstrated in-vitro on isolated gastric and duodenal cells clearly supports the idea that mucosal protection by PGs is a consequence of a direct action on PGs on the cells rather than manifestation of either a systemic or a CNS effects (21).

Table.3. Prostaglandins and Gastrointestinal Physiology.

Prostaglandins and Central Nervous System

Prostaglandins and other eicosanoids have been identified in the CNS. The synthesis of PGE2, PGD2, PGF2 alpha, PGI2, thromboxane A2 (TXA2), leukotriene C4 (LTC4), leukotriene B4 (LTB4) and other eicosanoids in the brain were well-demonstrated (Table 4; 22). Of interest is the observation that PGD2 and PGF2 alpha are synthesized in large quantity in the brain (22). PGD2 has recently been proposed as a mediator responsible for sleep (23). Both stimulant and depressant effects of PGs on the CNS have been reported following their injection into the cerebral ventricle and the firing rates of individual brain cells may be increased or decreased after iontrophoric applications of PGs (24). Intracerebroventricular administration of prostacyclin (PGI2) produced sedation, stupor, catatonia as well as cataleptic behavior (25). PGs have been proposed to modulate catecholaminergic (26), serotoninergic (27) and cholinergic (28) neurons in the CNS. There is also accumulating data suggesting possible modulatory role of PGs on dopamine mediated behavior (29). However, the evidence for a modulating role of PGs on neuronal pathways is derived from limited in-vitro studies and no studies have investigated the central role of PGs on peripheral dopamine-mediated GI effects.

Table.4. Prostaglandins and Central Nervous System

Convincing experimental data indicate that PGs function in mostly pathological processes in the CNS, including drug dependence, nociception, fever induction, learning and memory, and excitotoxic brain injury such as stroke and epilepsy (30, 31). Elevated levels of PGE2, PGF 2 alpha, thromboxane B2 in cerebrospinal fluid have been found in patients with AIDs dimentia. Abnormal central COX-2 expression had been found in patients with Parkinson’s disease and Down syndrome (32).

There is some emerging evidence indicating that central PGs may also be connected to an endogenous cannabinoids system as noted by the discovery that anadamide (arachidonyl ethanolamine), which is chemically an eicosanoid and is considered to possess cannabinoid agonist activity (33). However, it is beyond the scope of this paper to discuss all aspects of central PGs involvement in all pathological processes but rather focus on four principal actions of PGs as detailed below:
    (a) Drug dependence: The central administration of PGE2 facilitates acute dependence in morphine treated rats, while PGF2 alpha (acting on dopaminergic neurons) showed inconsistent attenuation of such dependence. Nielsen and Sparber (34) showed attenuation of morphine-induced withdrawal, while Nakagawa et al. (35) did not demonstrate any reduction of withdrawal symptoms following the intracerebral administration of PGF2alpha. Furthermore, Nakagawa et al. (35) showed that synthetic PGs acting on the prostaglandin EP3 receptor attenuated withdrawal jumping in morphine dependent mice, however, the intracerebral administration of PGF2 alpha showed no such effect. Nechifor et al. (36) showed that two metabolically stable chemical analogs of PGF2alpha, when administered intraperitoneally, reduced several symptoms of the withdrawal syndrome in rats with morphine-induced dependence. These observations suggest a role of central PGs in the induction of drug dependence, either directly or indirectly via a modulating effect on catecholaminergic, serotoninergic and cholinergic neurons in the CNS.
    (b) Pain: Central PGs are clearly involved in pain perception and induction (31, 37). As reported by Turnbach et al. (38), the intrathecal administration of PGE2 (1-100 nmol) or PGF2alpha (1-100 nmol) produced profound and dose-dependent mechanical hyperalgesia, but only weak thermal hyperalgesia and touched-evoked allodynia in rats. Both PGs produced dose-dependent increases in response of nociceptive specific neurons to mechanical stimuli (38).
    (c) Inflammation: Prostaglandin E2 is the major prostanoid produced centrally and in the periphery in animal models of acute and chronic inflammation, and its formation in both locations is blocked by COX-2 inhibitors (39). The PGE2-induced inhibition by COX-2 inhibitors in the brain may occur secondarily to peripheral action mediated by inhibiting local PGs formation, which elicit increased firing of pain fibers and consequent activation in PGs synthesis in the CNS (39).
    (d) Fever: The systemic administration of PGEs induces fever in laboratory animals and man via a CNS mediated mechanism of action (30, 40, 41). Pyrogens such as interleukin-1 (IL-1) act via hypothalamic release of PGs (42).

Prostaglandins and Brain-Gut Relationship

As of now, there are no meaningful studies, which characterized the in-vivo effect of central PGs on GI physiology such as acid secretion, GI motility and cytoprotection. Miura et al. (43) examined the receptor subtypes mediating the effects of PGE2 on parasympathetic preganglionic neurons that regulate the activity of pelvic visceral organs using neonatal rat spinal slices, in-vitro. These investigators showed that PGE2 increased the firing frequency to depolarizing current pulses, induced after discharges and inhibited spike potential after hyper-polarization but did not affect phasic preganglionic neurons. These results indicate that PGE2 acting via EP1 and/or EP4 receptors modulated the excitability and/or the excitatory synaptic input to tonic parasympathetic preganglionic neurons. Clearly, these studies need to be repeated in order to confirm the neurophysiologic action of PGE2 in the gut.

From a pathophysiologic considerations, prostaglandins may be also be involved in etiology of postoperative ileus as evident by increased spinal expression of COX-2 suggesting a primary afferent activation. This activation of primary afferents may subsequently initiate inhibitory motor reflexes to the gut, contributing to postoperative ileus (44).

Given the limited available CNS information, the function of PGs in neuronal tissues rests on inferences from in-vitro studies and from the studies connected with COX inhibitors (32). The absence of specific PG receptor antagonists for E, D, F and I series has clearly hampered our understanding of the role of individual PG in the CNS as well as other tissues. We fully agree with the assessment of Morrow and Roberts (24) that there is no clear physiological role of PGs in the CNS. Furthermore, despite some indirect and preliminary evidence summarized in Table 5, there is no consistent data, which well demonstrate the involvement of PGs in the brain-gut axis.

Table.5. Prostaglandins and Brain-Gut Axis

In summary, PGs and COX enzymes are present in and out of the CNS. Elevated levels of PG are found in few pathological processes such as fever and pain. The function of PGs in neuronal tissues rest on inferences from in-vitro studies and from studies connected with COX inhibitors. We conclude that the GI physiological and mucosal protective effects of PGs are essentially mediated by direct effects on cells or organs rather than by a direct effect on the CNS. Clearly, additional studies are warranted to investigate the CNS role in the GI physiological actions of PGs to clarify the precise role of PGs in brain-gut axis.


REFERENCES
  1. Dajani EZ: The actions of prostaglandins on the gastrointestinal tract. In: Prostaglandins in the Upper Gastrointestinal Tract. Vol. 5, in Therapeutics Today Series, Addis Press, Australasia Pty., Ltd., 1986, pp 9-18.
  2. Dajani EZ, Driskill DR, Bianchi RG, Collins PW, Pappo R. SC-29333, a potent inhibitor of canine gastric secretion. Dig Dis Sci 1976; 21: 1049-1057.
  3. Glavin G. Dopamine and gastroprotection. The brain-gut axis. Dig Dis Sci 1991; 36: 1670-1672.
  4. Pappas T, Tache Y, Debas H. Opposing central and peripheral actions of brain-gut peptides: A basis for the regulation of gastric function. Surgery 1985; 98: 183-190.
  5. Glavin GB, Hall AM. Brain-Gut relationships: Gastric mucosal defense is also important. Acta Physiol Hungar 1992; 80: 107-115.
  6. File S, Pearce J. Benzodiazepines reduce gastric ulcers induced in rats by stress. Br J Pharmacol 1981; 78: 593-599.
  7. Hernandez D, Xue B. Imipramine prevents stress gastric glandular lesions in rats. Neurosci Lett 1989; 103: 209-212.
  8. Dajani EZ, Bianchi RG, Calhoun DW. Synergistic actions of propantheline bromide with cimetidine and thiopropazate hydrochloride in the prevention of stress ulcer formation in rats. J Pharmacol Exp Therap 1979; 210: 373-377.
  9. Szabo S, Sandrock A, Nafradi A, Maull A, Gallagher G, Blyzniuk A. Dopamine and dopamine receptors in the gut. Their possible role in duodenal ulceration. Adv Biosci 1982; 37: 165-170.
  10. Szabo S, Horner H, Maull H, Schnoor J, Chiveh C, Palkovits M. Tissue catecholamines and serotonin in duodenal ulceration caused by cysteamine or prorionitril in the rat. J Pharmacol Exp Ther 1987; 240: 871-878.
  11. Bugajski J. Role of prostaglandins in the stimulation of the hypothalamic-pituitary-adrenal axis by adrenergic and neurohormone systems. J Physiol Pharmacol 1996; 47: 559-575.
  12. Bodis B, Karadi O, Nagy L, Dohoczky Cs, Kolega M, Mozsik Gy. Direct cellular effects of some mediators, hormones and growth factor-like agents on denervated (isolated) rat gastric mucosal cells. J Physiol (Paris) 1997; 91: 183-187.
  13. Dajani EZ, Roge EAW, Bertermann RE. Effects of E-prostaglandins, diphenoxylate and morphine on intestinal motility in vivo. Eur J Pharmacol 1975; 34: 105-113.
  14. Dajani EZ, Bertermann, RE, Roge EAW, Schweingruber FL, Woods EM. Canine gastrointestinal motility effects of prostaglandin F2 alpha in vivo. Arch Int Pharmacodyn Therap 1979; 237: 16-24.
  15. Rainsford KD, Peskar BM. Relationship between gastric ulceration and mucosal prostaglandins concentrations following chronic administration of aspirin preparations to pigs. Agents Actions 1979; Suppl. 4: 293.
  16. Krupp P, Menasse R, Riesterer L, Ziel R. The biological significance of prostaglandin synthesis. In: The Role of Prostaglandins in Inflammation. Lewis, Ed., Hans Huber, Bern, 1976, 106.
  17. Agrawal NM, Dajani EZ. Drug-induced ulcers. In: Gastrointestinal Pharmacology and Therapeutics. J Friedman, ED Jacobson, R McCallum (eds), Lippincott-Raven Publishers, Hagerstown, Maryland, Chapter 5, 1997, pp 55-63.
  18. Dajani EZ, Agrawal NM. Prevention and treatment of NSAID-induced ulcers: An update. J Physiol Pharmacol 1995; 46: 3-16.
  19. Dajani EZ, Callison DA, Bertermann RE. Effects of E-prostaglandins on canine gastric potential difference. Dig Dis Sci 1978; 23: 436-442.
  20. Dajani EZ, Nissen CA. Investigations of the gastrointestinal cytoprotective effects of misoprostol: Clinical efficacy overview. Dig Dis Sci 1985; 30: 194S-200S.
  21. Domschke S, Dembinski A, Domschke W. Partial prevention of ethanol damage of human gastroduodenal mucosa by prostaglandin E2 in vitro. Scand J Gastroenterol 1983; 18: 113-116.
  22. Wolfe LS. Eicosanoids: Prostaglandins, thromboxanes, leukotrienes, and other derivatives of carbon-20 unsaturated fatty acids. J Neurochem 1982; 38(1): 1-14.
  23. Urade Y, Hayaishi O. Prostaglandin D2 and sleep regulation. Biochim Biophys Acta 1999; 1436: 606-615.
  24. Morrow JD, Roberts LJ. Lipid derived Autacoids. Eicosanoids and platelet-Activating Factor. In: Goodman & Gilman’s The Pharmacological Basis of Therapeutics. JG Hardman, LE Limbird (eds). 10th Edition, New York, McGraw Hill; 2000, pp. 669-685.
  25. Brus R, Szkilnik R, Slominsk-Zurek J, Krzeminski T, Herman ZS. Studies on the behavioral and hypotensive effects of intraventricular Prostacyclin (PGI2) in rats. Polish J Pharmacol Pharm 1981; 33: 467-474.
  26. Narumiya S, Ogorochi T, Nakao K, Hayaishi O. Prostaglandin D2 in the rat brain, spinal cord and the pituitary: basal level and regional distribution. Life Science 1982; 31: 2093-2103.
  27. Brus R, Herman ZS, Szkilnik R, Zabawska J. Mediation of central prostaglandin effects by serotoninergic neurons. Psychopharmacology 1979; 64: 116-120.
  28. Saito R, Kamiya H, Ono N. Role of central muscarinic receptor of prostaglandin I2 in cardiovascular function in rat. Brain Res 1985; 330: 167-169.
  29. Naidu PS, Kulkarni SK. Differential effects of cyclooxygenase inhibitors on haloperidol-induced catalepsy. Progress in Neuro-Psychopharmacology & Biological Psychiatry 2002; 26: 819-822.
  30. Vidensky S, Zhang Y, Hand T et al. Neuronal over-expression of COX-2 results in dominant production of PGE2 and altered fever response. Neuromolecular Med 2003; 3(1) 15-28.
  31. Ito S, Okuda-Ashitaka E, Minami T. Central and peripheral roles of prostaglandins in pain and their interactions with novel neuropeptides nociceptin and nocistatin. Neurosci Res 2001; 41 (4): 299-332.
  32. Kaufmann WF, Andreasson KI, Isakson PC, Worley PF. Cyclooxygenase and the central nervous system. Prostaglandins 1997; 54: 601-624.
  33. De Petrocellis L, Melck D, Ueda N et al. Brain and peripheral anadamide amidohydrolase and its inhibition by arachidonate analogues. In: Recent Advances in Prostaglandin, Thromboxane and Leukotriene Research. H Sinzinger, B Samuelsson, JR Vane, R Paoletti, P Ramwell, PY-K Wong (eds), Plenum Press, New York 1998, pp. 259-263.
  34. Nielsen JA, Sparber SB. Central administration of prostaglandin E2 facilitates while F2 alpha attenuates acute dependence upon morphine rats. Pharmacol Biochem Behav 1985; 22 (6): 933-939.
  35. Nakagawa T, Minami M, Katsumata S, Ienaga Y, Satoh M. Suppression of the nalaxone-precipitated withdrawal jumps in morphine-dependent mice by stimulation of prostaglandin EP3 receptor. Br J Pharmacol 1995; 116(6): 2661-2666.
  36. Nechifor M, Chelarescu, Teslariu E, Cocu F, Negru A. Effects of PGF2 alpha analogs in experimental morphine-induced pharmacodependence. In Advances in Prostaglandin, Leukotriene and other Bioactive Lipid Research. Z Yazici, GC Folco, JM Drazen, S Nigam, T Shimizu (eds). Kluwer Academic/Plenum Publishers, New York, 2003, pp. 121-124.
  37. Dirig DM, Yaksh TL. Hyperalgesia-associated spinal synthesis and release of prostaglandins. In Recent Advances in Prostaglandin, Thromboxane and Leukotriene Research. H Sinzinger, B Samuelsson, JR Vane, R Paoletti, P Ramwell, PY-K Wong (eds). Plenum Press, New York 1998, pp. 205-208.
  38. Turnbach ME, Spraggins DS, Randich A. Spinal administration of prostaglandin E2 or prostaglandin F2 alpha primarily produces mechanical hyperalgesia that is mediated by nociceptive specific spinal dorsal horn neurons. Pain 2002; 97: (1-2) 33-45.
  39. Ciceri P, Zhang Y, Shaffer AF et al. Pharmacology of celecoxib in rat brain after kainate administration. J Pharmacol Exp Therap 2002; 302: 846-852.
  40. Milton A, Wendlandt S. A possible role for prostaglandin E1 as a modulator for temperature regulation in the central nervous system of the cat. J Physiol 1970; 207: 76P-77P.
  41. Bernheim HA, Gilbert TM, Stitt JT. Prostaglandin E levels in the third ventricular cerebrospinal fluid of rabbits during fever and changes in body temperature. J Physiol 1989; 301: 69-78.
  42. Coceani F, Lees J, Bishai I. Further evidence implicating prostaglandin E2 in the genesis of pyrogen fever. Am J Physiol 1988; 254: R463-R469.
  43. Miura A, Kawatani M, Mauyama T, De Groat WC. Effect of prostaglandins on parasympathetic neurons in the rat lumbosacral spinal cord. Neuroreport 2002; 13(12): 1557-1562.
  44. Kreiss C, Birder LA, Kiss S, VanBibber MM, Bauer AJ. COX-2 dependent inflammation increases spinal Fos expression during rodent postoperative ileus. Gut 2003; 52(4): 527-534.

R e c e i v e d:  November 15, 2003
A c c e p t e d: December 15, 2003

Author’s address: Esam Z. Dajani, Ph.D., FACG, IDDC Corporation, 1549 RFD, Long Grove, IL 60047-9532, USA, Phone (847) 634-9586, Fax (847) 634-3349
E-mail: EsamD@aol.com