Phosphate Depletion Associated With Hypoxia, Autonomic Neuropathy, Hypoventilation, or Paralysis: Potential Relevance to Sudden Infant Death Syndrome

These [Siddiqui and Bertorini, 1998: (http://www.ncbi.nlm.nih.gov/pubmed/9572247); Gravelyn et al., 1988: (http://deepblue.lib.umich.edu/bitstream/2027.42/27325/1/0000348.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/3364446); Steckman et al., 2006: (http://www.ncbi.nlm.nih.gov/pubmed/16642427); Heames and Cope, 2006: (http://www.ncbi.nlm.nih.gov/pubmed/17090245)] are some interesting articles that show some of the variegated manifestations of hypophosphatemia. A crucial fact that I've taken from the research on phosphate homeostasis (it's arguably the most crucial point) is that neither the steady-state nor the between-dosage (in the context of phosphate infusions in animals or phosphate supplementation in humans) intracellular phosphate levels, in either muscle cells or red blood cells, correlates with the serum phosphate levels. For example, Chobanian et al. (1995) [Chobanian et al., 1995: (http://www.ncbi.nlm.nih.gov/pubmed/7900836)] found that the intracellular ATP concentrations in cells in the proximal tubules correlated positively with the intracellular inorganic phosphate (Pi) concentrations, and artificially-induced changes in extracellular Pi concentrations produced changes in the intracellular Pi concentrations. But in human studies, the intracellular Pi values generally do not correlate with serum Pi values, and the intracellular Pi concentrations can be significantly depleted in a person who has a normal serum Pi level.

That usual absence of a correlation between intracellular and serum Pi concentrations means, in my opinion, that intracellular phosphate depletion, in "normophosphatemic" people, should be considered as a possible factor contributing to some of these conditions that have been associated with hypophosphatemia. Siddiqui and Bertorini (1998) cited research showing that phosphate depletion can produce neuropathy that mimics Guillain-Barre syndrome, and the authors described the symptoms of a patient who developed neurological symptoms after she had been given parenteral nutrition without phosphate. The manifestations of neuropathy were suggestive of demyelinating polyneuropathy but were rapidly reversed by phosphate supplementation, meaning that there wasn't demyelination. The authors also discussed the fact that an increase in hexokinase activity, in response to insulin that has been released after the intake of carbohydrates, is thought to be an important factor that mediates the carbohydrate-induced increase in the transport of phosphate into cells and the decrease in serum phosphate that can result from that transport (Siddiqui and Bertorini, 1998). The authors also cited research showing cognitive dysfunction and encephalopathy in hypophosphatemic or (merely) intracellular-phosphate-depleted people (Siddiqui and Bertorini, 1998). One interpretation of the article by Steckman et al. (2006), in which gallstone-induced pancreatitis occurred in conjunction with hypophosphatemia and improved in response to phosphate administration, is that the phosphate depletion was causing neuropathy and interfering with gallbladder contractions. Neuropathy is known to be associated with gallbladder disease, and the normal functioning and contraction of the gallbladder is regulated by its autonomic (and sensory) innervation [(http://scholar.google.com/scholar?q=neuropathy+gallbladder+gallstone&hl=en);
the visceral sensory innervation can influence mast cell degranulation in the gallbladder, via the efferent-action-potential-mediated release of neuropeptides, and changes in mast cell degranulation and neuropeptide release can influence the autonomic regulation of gallbladder functioning, etc.: (http://scholar.google.com/scholar?hl=en&q=%22mast+cell%22+gallbladder+CGRP+OR+%22substance+P%22+OR+%22vasoactive+intestinal+peptide%22)]. Another interpretation would be to say that the phosphate depletion caused ATP depletion in the liver and led to cholestasis, etc. Similarly, the respiratory muscle weakness found in association with hypophosphatemia or low serum phosphate levels (Gravelyn et al., 1988, cited above) could be a result of autonomic dysfunction, particularly given that hypophosphatemia can cause reversible quadriparesis (paralysis, meaning the people are transiently quadripalegics) (http://scholar.google.com/scholar?hl=en&q=quadriparesis+hypophosphatemia). The hypoventilation that can accompany hypophosphatemia could also be due to autonomic neuropathy and ATP depletion in parts of the brain (http://scholar.google.com/scholar?hl=en&q=hypoventilation+hypophosphatemia). Hypophosphatemia has also shown up in association with extrapontine myelinolysis (which is central "pontine" myelinolysis that doesn't occur in the pons, essentially), one of the forms of osmotic demyelination that can result from the excessively-rapid correction of hyponatremia with intravenous, hypertonic saline [Qadir et al., 2005: (http://www.jpma.org.pk//PdfDownload/759.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/16045098)]. The authors suggested that ATP depletion in glial cells in parts of the brain might have contributed to the case, but I'm not sure that the authors actually said that the phosphate might have contributed to or caused the ATP depletion. The intracellular phosphate may well have been depleted in parts of the brain, and that depletion may have impaired volume regulation and predisposed to the osmotic demyelination.

In any case, I found this article showing "sinusoidal" seasonal changes in the incidence of sudden infant death syndrome (SIDS) (the seasonal change in the incidence shows up in the Southern and Northern hemispheres, and SIDS was found to peak in the winter in both hemispheres) [Douglas et al., 1996: (http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2351134)(http://www.ncbi.nlm.nih.gov/pubmed/8646093)], and there's old research suggesting an association of SIDS with vitamin D depletion or differences in vitamin D metabolism or rickets, etc. [Schluter, 1996: (http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=2352183&blobtype=pdf)(http://www.ncbi.nlm.nih.gov/pubmed/8842097); (http://scholar.google.com/scholar?hl=en&q=%22sudden+infant+death%22+%22vitamin+D%22)] (or with other light-associated changes, such as involving changes in melatonin levels induced by sleeping on the back as opposed to the side, etc.) (Douglas et al., 1996). There's also research showing that infants who were experiencing apnea were more likely to be hypercalcemic than infants not experiencing apnea [Kooh and Binet, 1990: (http://www.pubmedcentral.nih.gov/picrender.fcgi?artid=1452283&blobtype=pdf)(http://www.ncbi.nlm.nih.gov/pubmed/2207905)]. I couldn't get results to show up on a quick search, but hypercalcemia has been found to occur in hypophosphatemic people. Although Kooh and Binet (1990) didn't find that serum phosphate levels were associated with apnea in any way, the serum Pi levels wouldn't have to. Given that intracellular Pi levels do not reliably correlate with serum Pi levels and that phosphate depletion is known to be capable of causing respiratory paralysis/hypoventilation and hypoxia and neuropathy [see this and many others, some of which I discussed above: Weber et al., 2000: (http://www.ncbi.nlm.nih.gov/pubmed/10663486)] [and given that vitamin D depletion is known to be a cause of phosphate depletion (and that vitamin D supplementation, even in the absence of any genetic defect specifically involving vitamin D receptor signalling)], one possibility is that intracellular phosphate depletion in parts of the brain (and in the red blood cells, causing low-level hypoxia that might gradually have more severe consequences) could contribute to some cases of SIDS. Although there was one small study showing no apparent depletion of 25-hydroxyvitamin D levels in the context of SIDS, there could very easily be different degrees of intracellular phosphate depletion among infants with the same 25(OH)D levels. And looking at the serum phosphate levels wouldn't necessarily show anything, given the lack of correlation of intracellular and serum Pi levels. Someone would have to use MRS scans or look at the intracellular 2,3-DPG or Pi levels in red blood cells in infants, instead of just looking at the serum Pi. It's interesting that Heames and Cope (2006) (cited above) found that they could reduce the rate of infusion of noradrenaline in a manner that was proportional to the increase in serum phosphate, in a person who had developed transient heart failure from postsurgical phosphate depletion. The phosphate depletion basically caused hypotension, and the interactions with noradrenaline are really interesting (the usual thing people discuss is the fact that adrenergic drugs decrease serum phosphate by promoting phosphate uptake into cells). Given the changes in the autonomic regulation of blood pressure that occur in response to changes in the orientation of the body, such as in a baby sleeping prone vs. supine (http://scholar.google.com/scholar?q=autonomic+orthostatic+prone+supine&hl=en), it's possible that there's a kind of feed-forward depletion of intracellular phosphate in parts of the brain that can lead to apnea and then increased ventilation to compensate (and then phosphate depletion because of that and because of the noradrenaline released in response to that, as in the stress response to hypoxia, and to the potential vitamin D-depletion-induced renal phosphate wasting, etc.).

Arguably, the most well-established cause of hypophosphatemia is alkalosis induced by hyperventilation (http://scholar.google.com/scholar?q=hyperventilation+alkalosis+hypophosphatemia&hl=en), and apnea commonly occurs in response to post-hyperventilation alkalosis (http://scholar.google.com/scholar?q=hyperventilation+apnea&hl=en). So the alkalosis, in response to hyperventilation (as in response to autonomic dysfunction during sleep, resulting from changes in the sleep position and from phosphate depletion in neurons or smooth muscle cells or muscle cells in the diaphragm), could drive phosphate into cells outside the brain, thereby reducing phosphate availability to the brain, and then that could gradually set the stage for more severe episodes of hypoxia, more autonomic dysfunction due to the phosphate depletion in the brain, etc. There's evidence of repeated episodes of hypoxia in some research on SIDS [see Takashima et al. (1978) and Rognum et al. (1991): (http://scholar.google.com/scholar?hl=en&q=%22sudden+infant+death%22+hypoxia)]. A decrease in the responsiveness of smooth muscle cells (or other cell types, as in neurons in the brainstem, in the context of phosphate depletion) to noradrenaline occurs in people who have orthostatic hypotension and other derangements of baroreceptor functioning, and L-threo-3,4-dihydroxyphenylserine (DOPS) has been researched as a treatment for orthostatic hypotension and orthostatic tachycardia (DOPS is a precursor of noradrenaline) (http://scholar.google.com/scholar?hl=en&q=orthostatic+DOPS). Hypophosphatemia has been associated with instability in blood pressure, in association with postural hypotension and other problems with the sensitivity and functioning of the baroreceptor reflexes (http://scholar.google.com/scholar?hl=en&q=orthostatic+hypophosphatemia). Anyway, I just put those types of crude thoughts up on this blog.

Free-Wheeling Discussion of L-Methylfolate

In this article [Di Palma et al., 1994: (http://cat.inist.fr/?aModele=afficheN&cpsidt=4093099)], Di Palma et al. (1994) used 90 mg/day of methylfolate to treat depression and discussed research on the use of 50 mg/day of methylfolate [including Guaraldi et al., 1993: (http://www.ncbi.nlm.nih.gov/pubmed/8348200)]. There's more research on the use of 50 mg per day, but I think some are abstracts. Di Palma et al. (1994) also keep repeating that the patients who participated in those trials were "normofolatemic" (i.e. displayed normal serum folate levels). The main issue I would wonder about is the use of supplemental folic acid per se, past a certain dosage. It seems to cause some strange effects, at dosages above 10 or so mg/day, in the long term. It's not that it's really toxic [except in people who have dihydropteridine reductase deficiency, an inherited genetic disorder (http://hardcorephysiologyfun.blogspot.com/2009/05/evidence-that-reduced-folates-can-serve.html)] but that it may compete with methylfolate for entry into the brain (as suggested by the authors of one of those articles on dihydropteridine reductase deficiency) and may compete with tetrahydrobiopterin (BH4) for binding to tyrosine hydroxylase or the nitric oxide synthases. But if one is not taking supplemental folate or has a serum folate value that is within the normal range (due to some dosage of folic acid below 5 mg or something), I don't think the serum folate level is really relevant to the effects of methylfolate. The serum folate range is very small, and a serum folate level within the normal range of values is unlikely, in my opinion, to produce anything close to "saturation" of the intracellular binding sites for the intracellular total folates (the concentration of binding sites, in the liver, is something like 150-200 uM, at least, excluding the binding sites on the pterin biosynthetic enzymes). But one would want to discuss this type of thing with one's doctor, especially if one were taking any medications. In that past posting I linked to, the authors of one of the articles go into all the research on the use of reduced folates (such as methylfolate) and oxidized folates (folic acid) in people who have dihydropteridine reductase deficiency, and the authors provided a lot of evidence that reduced folates tend to actually be safer than folic acid, in terms of their effects on the brain. That seems paradoxical, at first glance, because reduced folates are more potent, from the standpoint of their effects on cell proliferation, etc. But they exert less of a pro-convulsant effect, for example, and don't cause the neurological symptoms that folic acid does in people with dihydropteridine reductase deficiency (DHPRD) and in some forms of phenylketonuria, I think, in which reduced folates have been used instead of BH4 (this was before BH4 was approved to treat BH4 depletion in phenylketonuria). That's probably because of the lack of capacity of folic acid to serve as a BH4 analog, except to the extent that some of it can be reduced (converted into reduced folates intracellularly).

One thing I can think of that would be a downside of high dosages of methylfolate would be the potential to mask vitamin B12 deficiency, and I think there's reason to consider dosages of methylcobalamin in the 1-5 mg/day range, in combination with methylfolate, even assuming that the serum B12 is normal. The other thing is that methylfolate can probably serve as a BH4 analog in humans (it seems to, in my opinion), and, to the extent that it can, that could conceivably cause some sort of abnormal nitrergic effects at high dosages. That doesn't seem to be much of a downside, in my opinion, although a person who had an inflammatory disease or something like multiple sclerosis could experience some sort of mixture of bad and good effects from methylfolate at high dosages, as a result of the BH4 "mimesis" at higher dosages. But the supposed nitrergic effect is also likely to contribute to the dopaminergic/noradrenergic effects of methylfolate, given that, for example, L-arginine releases dopamine in a BH4-dependent manner (although it's more complicated than one might think) [I'm pretty sure it's the Liang et al. (1998) article: (http://scholar.google.com/scholar?hl=en&q=%22L-arginine%22+dopamine+tetrahydrobiopterin)]. So it's not really likely to be a bad effect, up to a certain dosage (I mean that different people might experience differing degrees of nitrergic effects from methylfolate, and those effects might be problematic in some people and not in others, etc.). I tend to think that combining methylfolate with L-arginine might cause undesirable side effects, at high dosages of either one, and I tend to think the nitrergic effects of methylfolate in the brain (the supposed nitrergic effects) are likely to be "better" than those of arginine. L-arginine just seems to produce inconsistent effects or to produce a mixture of mood-elevating and mood-worsening effects. It's not a matter of toxicity, but it just doesn't seem to produce very predictable effects. The research shows that inconsistency.

The main issue, in my opinion, would be to exercise some caution in combining high dosages of methylfolate with noradrenergic or dopaminergic medications, because the BH4-mimicking effect of methylfolate would be expected to augment noradrenergic/dopaminergic effects. It's not that it's really about neurotoxicity, in my opinion, but that methylfolate could just augment the effects of those medications and cause agitation or insomnia or nervousness, etc. That's the main effect of BH4. Its predominant effect is to enhance dopaminergic and noradrenergic transmission in the brain, and methylfolate really seems to begin serving as a BH4 analog at high dosages. For example, they've used BH4 to treat dopa-responsive dystonia [(http://scholar.google.com/scholar?hl=en&q=%22dopa+responsive%22+dystonia+tetrahydrobiopterin+supplement); (http://scholar.google.com/scholar?hl=en&q=%22dopa+responsive%22+dystonia+tetrahydrobiopterin)], and I think that's mainly or always caused by mutations affecting BH4 biosynthesis. I can't say that methylfolate definitely serves as a BH4 analog at higher dosages, and I've never taken BH4 and therefore can't make even a subjective comparison (BH4 has been used to treat depression, in a number of small studies, is available by prescription in the US, and is sold under the generic name of sapropterin dihydrochloride; but it's not available over-the-counter and seems unlikely to be something that would be covered by prescription). L-methylfolate is available by prescription or over-the-counter, as discussed in past postings.

I tend to think that the way methylfolate works is to cause something resembling saturation of the binding sites in the liver (and maybe also the brain or other extrahepatic tissues), at the "lower" range of dosages (I mean that this quasi-saturation might begin to emerge at, say, 15 or 20 mg/day or maybe less), and then begin to serve as a BH4 analog at higher dosages. In my experience, the effects of L-methylfolate are greater at the higher dosage range, but there's some dosage range at which one would experience diminishing returns from further increases in dosages. But Di Palma et al. (1994) found, incidentally (without seeking to find any such effect), that the patients' liver enzymes were decreased by the 90 mg/day dosage. If methylfolate were toxic at that dosage, one wouldn't expect to see a reduction in serum liver enzyme levels (the reductions were significant, and I don't have the article in front of me right now) and might expect to see a worsening of liver function (given that the intracellular total folate levels in the liver would be higher than those in any other tissue). But it basically improved liver function, to some extent, in those patients. It's never been proven to treat liver disease or any other disease, however, but I'm just saying that that suggests that the higher dosages are unlikely to be "toxic." I can't make any definitive statements about safety, however, on an individual basis, and one would want to discuss that type of thing with one's doctor. Di Palma et al. (1994) didn't think 90 mg/day was toxic and mentioned that, but they didn't necessarily have any basis for saying that (other than their clinical experience in psychopharmacology). But anyone with any health condition would obviously want to be extra careful and discuss the matter in more depth with one's doctor. For example, low dosages of reduced folates have sometimes produced anticonvulsant effects in people who have specific, probably-genetic forms of epilepsy due to cerebral folate deficiency, but higher dosages could conceivably lower the seizure threshold in the way that many antidepressants do (as far as I know, antidepressants tend to lower the seizure threshold, almost without exception). In general, in my opinion, L-methylfolate is useful (and much more useful than folic acid).

Tricyclics as Lipophilic, Weak Bases: Relevance (or Irrelevance) to Supposed Uncoupling Effects in Mitochondria

The authors of this article [Stott et al., 1996: (http://cat.inist.fr/?aModele=afficheN&cpsidt=3197621)] discuss research showing that the pKa values for the secondary amine nitrogens of doxepin, amitriptyline, and imipramine range from 9.0 to 9.4. The tricyclics' uncoupling effects, to the extent that they may occur in vivo (http://hardcorephysiologyfun.blogspot.com/2009/07/mild-uncoupling-as-factor-that-may.html), may be at least partially a result of the tricyclics' being lipophilic weak bases and also roughly planar, etc. Some uncouplers are lipophilic, weak bases with pKa values in that range (others are weak acids) (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=pKa+mitochondrial+uncoupling+%22weak+base%22+OR+%22weak+acid%22); (http://hardcorephysiologyfun.blogspot.com/2009/05/some-structure-activity-relationships.html)]. For example, Garlid et al. (1983) [Garlid et al., 1983: (http://www.jbc.org/cgi/reprint/258/13/7974.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/6223029)] noted that lipophilic amines can produce uncoupling and mimick the effects of weak acid uncouplers (Garlid et al., 1983). Some lipophilic amines, as Garlid et al. (1983) discuss, can produce uncoupling by interacting with anions in "ion pairs." It's interesting that Garlid et al. (1983) found some uncoupling effects from quinidine, because tricyclics, especially at higher or toxic doses, produce quinidine-like cardiotoxicity, partly as a result of their anticholinergic effects, and thereby produce QT-interval prolongation and hence produce quinidine-like anti-arrhythmic effects [(http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=quinidine+tricyclic); (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=quinidine+tricyclic+antiarrhythmic)]. Garlid et al. (1983) also tested some other drugs, such as chlorpromazine, used in psychiatry, and some local anesthetics. Some of those anesthetics have overlapping pharmacological mechanisms with tricyclics (sodium-channel blocking effects), but that could just mean that the different drugs are lipophilic amines and weak bases. But there does tend to be a structural requirement, such as a bulky, aromatic substituent, for a drug to produce uncoupling. It's not enough that it's a weak base. Its half-life, for example, would be one determinant. I think tacrine accumulates, for example, over time, more than many drugs (the authors discuss that in the article I cited in the previous posting (http://hardcorephysiologyfun.blogspot.com/2009/07/mild-uncoupling-as-factor-that-may.html).

The "bulky," heterocyclic ring system probably would play a role in any supposed uncoupling effects, also, but I'm not sure what the role would be. One might ask why not all antidepressants are weak acids or weak bases and why not all of them produce uncoupling, etc. I'm not trying to explain antidepressant actions in general. This is just one factor that seems like it might be relevant and might be one effect of some antidepressants. Obviously, there would be potential for harm from uncoupling, but the point is that the induction of mild uncoupling by some of these approaches (even exercise may have some of those effects in astrocyte mitochondria) could be viewed as being a bioenergetic compromise of some kind. Another possibility is that the wastefulness of it, the ATP depletion, could cause an adaptive change in glycogen storage, etc.

It's interesting that reserpine (a plant-based drug that depletes vesicular stores of monoamine neurotransmitters, by blocking their uptake into vesicles by vesicular monoamine transporters, and is basically the opposite of an antidepressant) produces uncoupling (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=uncoupling+mitochondria+psychiatric+OR+psychoactive+OR+antidepressant), and that could either cast serious doubt on this concept or could indicate that severe uncoupling produces severe ATP depletion that the stimulation of respiration, in response to the uncoupler, can't compensate for. In any case, I still think it's interesting that chlorpromazine was the strongest uncoupler in some of the experiments that Garlid et al. (1983) conducted, and that might suggest that there can be a spectrum of effects. Milder uncouplers might stimulate respiration enough to more than compensate for the initial ATP depletion, produced as a result of the futile cycling that the uncoupler causes. Stronger uncouplers might produce more significant ATP depletion and "tranquilizing" or toxic effects, etc.

In any case, those are interesting articles, but these are just inchoate thoughts. It's interesting to think about that type of thing.

Mild Uncoupling as a Factor That May Mediate Antidepressant Effects of Tricyclics and Other Compounds

This article [Berson et al., 1996: (http://www.ncbi.nlm.nih.gov/pubmed/8964414)] discusses the capacity of tacrine to produce uncoupling in mitochondria and to increase respiration, and I was noticing the crude structural similarity between tacrine and some of the tricylic antidepressants (http://scholar.google.com/scholar?q=tacrine+imipramine+tricyclic&hl=en&safe=off&num=100&um=1&ie=UTF-8&oi=scholart). Tacrine is sort of an extreme example, in my opinion, but some tricyclics also can produce some slight uncoupling effects in mitochondria (i.e. of redox reactions from oxidative phosphorylation) [see, for example, references in Wong et al., 2004: (http://149.142.238.229/pgxlab/docs/Wong-SJW-2004-MP.pdf)(http://www.ncbi.nlm.nih.gov/pubmed/14743185)], and hypericin (which is a planar molecule that, in the sense that a planar ring system is one structural feature of some uncouplers, is structurally "similar," in a very general sense, to tricyclics), for example, can either facilitate or inhibit uncoupling in mitochondria (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=hypericin+mitochondrial+respiration). Some of the articles on uncoupling that I've cited in past postings provide better discussions of planar, "heterocyclic" ring systems being a structural feature of uncouplers, but here are some references (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=planar+uncoupling+mitochondria+orbital). I'm having trouble getting full texts of these types of articles, but I think the tricyclics can either assume planar or stacked-planar conformations that may explain their uncoupling effect [Maxwell et al., 1970: (http://jpet.aspetjournals.org/cgi/content/abstract/173/1/158)]. I forget what the conformational flexibility of that type of heterocyclic ring system, with the seven-carbon ring, is like. (In any case, it's not even really relevant to the discussion.)

Uncoupling is not always bad and is thought to be protective under many sets of circumstances. For example, uncoupling can protect against the generation of reactive oxygen species. I wonder if mild uncoupling contributes to the antidepressant effects of some tricyclics or to the St. John's Wort (SJW) extract. People may well have discussed this possibility in some articles, and I haven't really looked into it. The main problem some people have with SJW extract, which has considerable evidence to support its effects or usefulness or whatever, as an adjunctive antidepressant approach, is the photosensitivity. And I should mention that extended, planar ring systems, in compounds, are thought to mediate the photosensitizing effects of many compounds. There's actually a surprising amount of evidence suggesting that rutin, which I think is not photosensitizing and is sold separately, may contribute to the supposed antidepressant effects of SJW extracts (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=rutin+antidepressant). That research on rutin actually looks kind of questionable to me, and I'm inclined to doubt it does actually contribute to the effects of SJW. But that's just my opinion, and I put these types of things on the blog (as I come across them). I have no idea about that.

But Wong et al. (2004) note that, in animals, tricyclics have sometimes been shown to initially (within the first 7 days) increase glucose uptake or oxidation in the brain initially, consistent with a stimulation of respiration/oxygen uptake, resulting from a mild uncoupling effect. Subsequently, after 28 days, the rate of glucose oxidation was decreased, in response or as an adaptation to the presence or higher concentration of the tricyclic (reference 49, discussed on page 11). I wonder if that might partly explain the delay in the onset of the effects of tricyclics, etc.

These are crude thoughts, but the initial effect of an uncoupler on ATP production tends to be sort of neutral, from what I can tell. For example, the research on the apparent and mild uncoupling effects of beta-hydroxybutyrate or acetoacetate or sodium butyrate infusions in humans or animals, discussed in a fairly recent posting, shows that there can be a slight, initial decrease in the rate of ATP production in response to the presence of an uncoupler (in response to the proton cycling by the protonated form of the short-chain fatty acid, although it's not clear that ketones or short-chain fatty acids produce true uncoupling--they appear to produce something similar to it or may produce it indirectly). Then, there is a stimulation of respiration, in response to that initial effect, and an increase in oxygen uptake (producing the thermal effect, or thermogenic effect that is observed in response to i.v. beta-hydroxybutyrate infusion, etc.) that tends to offset some of the pH changes (at least in response to ketone infusion) and other initial effects of the mild uncoupling. That's thought to account for the increase in glucose utilization. Maybe there's some sort of adaptation that contributes to antidepressant effects. Creatine, for example, stimulates respiration, although it's not really an uncoupler, as far as I know. It's sort of the opposite. But maybe it's not so much about uncoupling vs. the inhibition of uncoupling. Maybe there needs to be some sort of shock to the status quo, in terms of state III vs. state IV respiration, such that drugs or compounds can produce opposite effects initially but lead to the same net, adaptive increase or decrease in respiration (the animal research seems to suggest that tricyclics can produce some adaptive decrease in respiration, almost in a way that's reminiscent of the effects of an increase in ketone oxidation by neurons or astrocytes). Tianeptine, for example, actually increases serotonin reuptake and has been used as an antidepressant (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=tianeptine+%22serotonin+reuptake%22), and the serotonin reuptake inhibitors are obviously one class of antidepressants. That's not much of an argument and is not evidence of much of anything, for many reasons, but the point is that one can, in some instances, arrive at the same net effect (among different individuals or even in the same individual) by administering either two treatments or compounds that have more or less opposite effects, etc.

In any case, an increase in noradrenergic transmission, such as would occur in response to tricyclics, also tends to stimulate oxidative metabolism in astrocytes [this is a really crude search: (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=noradrenergic+glycogen+astrocyte+respiration+OR+oxidation)]. So the effects of uncoupling could work in concert with the supposed astrocyte-glycogenolytic (and subsequent rebound increase in glycogen formation, as a phenomenon that displays a crude similarity to glycogen supercompensation) effect of an increase noradrenergic transmission, and the uncoupling effect and noradrenergic mechanisms would not be mutually exclusive. There's actually some evidence to suggest that there is a kind of glycogen supercompensation in the brain, in astrocytes, in response to glycogen depletion (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=%22glycogen+supercompensation%22+brain). In any case, creatine can increase glycogen storage/supercompensation in skeletal muscles (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=creatine+glycogen) and has been shown to produce antidepressant effects at low dosages (i.e. 3 grams/day or less) in three small studies (discussed in previous postings). In this search, one can see additional studies in which creatine's effects on "mood" were evaluated in a sort of informal or subjective way (http://scholar.google.com/scholar?num=100&hl=en&lr=&safe=off&q=creatine+monohydrate+antidepressant+OR+mood). There are other lines of evidence, but some of those articles are interesting.

Serum Uric Acid, Energy Metabolism, Sympathetic Activation, and Goal-Oriented Behavior or "Grant-Money-Getting" Behavior

This article [Hunter et al., 1990: (http://www.ncbi.nlm.nih.gov/pubmed/2345757)] is one that I cited in a previous posting (http://hardcorephysiologyfun.blogspot.com/2009/03/interactions-of-caffeine-with-purine.html), but I didn't have time to discuss some interesting research that Hunter et al. (1990) discuss and cite. There's old research showing positive associations of serum uric acid (UA) levels with goal-oriented behavior and, essentially, activity level in general, and there's also some more recent research looking at UA per se as being a supposedly-reliable mediator of hyperactivity or mania or whatever other conditions. What interests me is the reasons why UA might be associated with goal-oriented behavior. I remember that a professor I took a class from once mentioned research showing that higher serum UA levels were associated with more success in getting grant money (in successfully getting grants awarded, etc.). He was referring to old research, from the 1950's or 1960's, but I wonder if it doesn't have to do with brain activity in some generalized sense. There's a vast amount of research showing that electrical stimulation or glutamatergic stimulation or noradrenergic activity increases extracellular-fluid (ECF) adenosine and UA levels in the brain. It's a generalized response that may just have to do with an increase in the metabolic demands of neurons. I tend to think that the cerebral metabolic activity or noradrenergic activity might just be making people slightly more aggressive or driven and might be accompanied by increases in sympathetic outflow from the central nervous system, and that could account for the serum UA elevations. That type of process, however, would not mean that low serum UA levels could not also be associated with excessively-prolonged increases in noradrenergic activity and sympathetic activation. There could be a pathological activation that would eventually compromise beta-adrenergic sensitivity, such as in people with multiple sclerosis (in whom the serum UA levels tend to be very low). Astrocytic beta2-adrenoreceptor density and sensitivity has been reported to be very low in people with MS, and there's research associating prescriptions for asthma (specifically beta2-adrenoreceptor agonists) with lower incidences of MS. Obviously, taking beta-agonists would be potentially dangerous for people with MS, and one would want to discuss that type of thing with one's doctor. The association only was found when researchers looked at medical records across many years, also, although beta2-adrenoreceptor activation does tend to be anti-inflammatory and immunosuppressive. One could make the argument that robust increases and equally-robust decreases in noradrenergic activity in the brain, accompanied by augmentations in sympathetic outflow, would produce elevations in UA that would account for the associations of high serum UA with goal-oriented behavior. Poorly-regulated noradrenergic activity could conceivably lead to gradual, "functional sympathectomy-like" changes (reduced beta-adrenoreceptor sensitivity) that could produce decreases in serum UA, etc. This is very general and imprecise, but it's interesting to think about. Low serum UA is a generalized feature of a variety of intracranial disease states and is thought to be partially a result of poor osmoregulation in the brain, such that the sympathetic innervation of the kidneys changes. The decreases in functional, sympathetic innervation of the kidneys is thought to play more of a role in the etiology of cerebral salt wasting (CSW) than in the etiology of syndrome of inappropriate antidiuretic hormone secretion (SIADH). But there must be some more precise neurobiological changes that could account for the UA depletion that occurs in SIADH and CSW, and I'm not convinced that it only has to do with osmoregulatory failures per se or with changes in renal UA reabsorption. I think it might have to do with derangements in energy metabolism. Here's an interesting article that shows that the intramitochondrial UA levels are higher in rats with diabetes [Kristal et al., 1999: (http://www.ncbi.nlm.nih.gov/pubmed/10496973)]. But as the disease process and poor glycemic control in the rats' diabetes became more pronounced, the UA production normalized or decreased again. That's potentially really important for understanding why UA is low in people with MS, and it also casts serious doubt on the use of all of these association studies showing UA to be some kind of "independent" risk factor for (or variable independently-associated with) cardiovascular disease. One could claim to be able to control for insulin sensitivity in some association study, but that's unlikely to be possible. Energy metabolism, as related to insulin sensitivity, is far too complex to control for in an association study that looks at some blood tests from 20,000 people.

Sympathetic Activation During Resistance vs. Endurance Exercise: Relationship to Exercise-Induced Increases in Noradrenergic Transmission in the Brain

This article [Iellamo et al., 2002: (http://circ.ahajournals.org/cgi/reprint/105/23/2719)(http://www.ncbi.nlm.nih.gov/pubmed/12057984?dopt=Abstract)] is interesting, and it gives the reader at least some sense of the capacity of high-intensity exercise to produce quite different cardiovascular effects than low-intensity, aerobic exercise produces. There's a popular notion that aerobic exercise is the "heart-healthy" form of exercise and that weight training/resistance exercise is just about "muscle-building" or who knows what. The article by Iellamo et al. (2002) shows that high-intensity exercise tends to increase beta-adrenoreceptor responsiveness more than lower-intensity exercise, but the authors seem to evidently still be under the impression that this is "bad." There are just lots of problems with research in exercise. I know it's apparently impossible to research the effects of weight training in rodents, because they can only run on wheels, etc. But researchers keep doing studies on people who are in really pretty decent shape to begin with and then finding no changes or minimal changes in variables related to adrenergic functioning. In general, in my opinion, based on the information from many articles, resistance exercise produces more of an enhancement of beta-adrenoreceptor sensitivity but does not cause some kind of "pressor" effect, and endurance exercise is well-known to produce more of a pronounced increase in vagal (vagus nerve), or parasympathetic, tone, and this is not always purely beneficial to people. Sigal et al. (2004) [Sigal et al., 2004: (http://care.diabetesjournals.org/cgi/content/full/27/10/2518)(http://www.ncbi.nlm.nih.gov/pubmed/15451933)] discuss the fact that, during aerobic (endurance, lower-intensity) exercise, the regulation of glucose availability is primarily driven by these sort of subtle changes in neuroendocrine activity, and endurance exercise tends to decrease insulin levels and cause either unchanged or, actually, decreased plasma glucose levels, during exercise. In contrast, the increases in free fatty acids and plasma glucose that occur during resistance exercise are driven primarily by strong, sympathetic activation and can elevate adrenaline (epinephrine) levels by 15-fold, significantly elevate plasma growth hormone (GH) levels, and increase cortisol meaningfully, etc. So resistance exercise has a very different effect, and, in my opinion, "high-intensity" endurance exercise is not going to mimic those effects very effectively, if at all, in the long term. Reading various articles, one would think that no one has any problem with diminished sympathetic (I'm referring to adrenergic) tone, in terms of the vasoconstriction that is required for venous return to the heart, but this article discusses the high degree of prevalence of postural tachycardia (syncope can result from severe postural tachycardia/postural hypotension, and they're really talking about postural tachycardia and orthostatic intolerance in this article) [Van Lieshout et al., 2003: (http://jap.physiology.org/cgi/reprint/94/3/833)(http://www.ncbi.nlm.nih.gov/pubmed/12571122?dopt=Abstract)]. The authors discuss the fact that improving the strength of leg muscles can improve these symptoms significantly (postural tachycardia manifests itself as dizziness or as an inappropriate and prolonged increase in heart rate upon standing, etc.) and that the venous return to the heart plays a role in maintaining cerebral blood flow, etc.

It's well-known that pilots who fly some types of aircraft experience high G-forces can experience blackouts or "grayouts" tunnel vision, because of transiently diminished cerebral blood flow, and it's well known that resistance training, much more than endurance training, can reduce these symptoms by improving venous return, and Van Lieshout et al. (2003) mention some of that. The baroreflex that normally prevents orthostatic tachycardia or hypotension has multiple components and is really complex, but it's discussed, in much of the literature, as if everyone will benefit from an increase in vagal tone and that the vagal component of the baroreflex is the only relevant one. It's not, and the adrenergic activation that occurs in the brain is likely to be substantially more pronounced, in my opinion, during resistance exercise than during aerobic exercise.

The firing of muscle spindle afferent neurons increases during exercise and contributes to the activation of sympathetic neurons in the medulla, in the brainstem, and higher plasma adrenaline levels (which are obviously much higher during resistance exercise, in general, than during endurance exercise) correlate positively with larger amounts of noradrenaline release in the prefrontal cortex, in exercising rats [Pagliari et al., 1995: (http://www.ncbi.nlm.nih.gov/pubmed/7665408)]. The increases in plasma epinephrine result primarily from increases in the sympathetic outflow from the brain. The direct sympathetic innervation of the adrenal medulla allows epinephrine to be rapidly released during high-intensity exercise. ACTH is also released from the anterior pituitary gland during exercise and stimulates the release of cortisol from the adrenal cortex, during high-intensity exercise. Resistance exercise can induce prolonged elevations in plasma cortisol, at essentially all times during the increases and decreases in plasma cortisol that normally occur throughout the day. This can gradually contribute, over days, to an upregulation of beta-adrenoreceptor density and responsiveness and lead to changes in the magnitude and effects of the acute, exercise-induced increases in plasma epinephrine. Increases in plasma pCO2, during exercise, may also contribute to the activation of noradrenergic neurons in the A1/A2 adrenergic cell groups and the locus ceruleus, in the brain, during exercise [Bailey et al., 2003: (http://www.ncbi.nlm.nih.gov/pubmed/14513913)].

The release of noradrenaline in the prefrontal cortex is almost certainly a result, mainly, of increases in the firing rates of noradrenergic neurons whose cell bodies are in the locus ceruleus, although the other noradrenergic cell groups probably contribute more indirectly to that effect. The central noradrenergic activity, during exercise, is also required for brain-derived neurotrophic factor production in response to exercise, in the brain [Ivy et al., 2003 : (http://www.ncbi.nlm.nih.gov/pubmed/12759116), cited and discussed here: (http://hardcorephysiologyfun.blogspot.com/2008/12/noradrenergic-regulation-of-bdnf.html)], etc. There's a popular belief that increases in plasma beta-endorphin levels, which can occur during exercise, produce an opioidergic or morphine-like effect on the brain, etc., but an increase in plasma beta-endorphin levels is really an indication of a generalized stress response [Farrell et al., 1982: (http://www.ncbi.nlm.nih.gov/pubmed/7096149)]. To produce some kind of opioidergic reward, opioid peptides released into the blood would have to cross the blood-brain barrier, back into the brain, and somehow act selectively on opioidergic pathways that are involved in the mesolimbic reward system, and this isn't really likely to occur. There's a great deal of evidence that the increase in noradrenergic transmission, in the brain, during exercise is more likely to be a major factor regulating the subjective effects or mood elevation in response to exercise. Exercise can also increase dopamine release in the striatum (http://scholar.google.com/scholar?num=100&hl=en&lr=&q=exercise+dopamine+release), and the catecholaminergic effects of exercise are likely to be crucially important for many of its effects.

Another Old Paper of Mine

Writing this paper was an intellectual exercise, and I want to say that I strongly advise people to talk to their doctors before exposing themselves to sunlight or other UVB or UVA sources, even casually. And, in my opinion, it is not wise to use UVB or UVA for therapeutic purposes of any kind. I obviously think that sun exposure would increase the risk of skin cancers. Another implication of the mechanisms I discuss in the paper is that UV exposure could produce actual brain damage, and there are reports of people with multiple sclerosis dying after sitting in the sun. I think it would be a terrible idea for people to use UV therapeutically. In fact, UV exposure could cause brainstem hemorrhages by inducing low-frequency, spontaneous firing of trigeminal ganglion neurons, innervating the skin of the face. The low-frequency firing rates that primary afferent neurons demonstrate within ~30 minutes of exposure to UV causes CGRP and substance P to be released in the spinal cord and trigeminal nucleus, in the brain, and this could cause perivascular mast cell degranulation and thereby increase vascular permeability and cause a hemorrhage, etc. A single UV exposure can cause ~1 Hz, asynchronous firing of sensory neurons and trigeminal ganglion and dorsal horn neurons that lasts 5-7 days. But CGRP has immunomodulatory effects and could act on astrocytes and microglia, upon its release in various parts of the brain, etc. Another implication is that UV could modify cerebral blood flow (either decreasing it or increasing it) by inducing one or another type of trigeminal root reflex, meaning efferent action potentials in fibers innervating the cerebral blood vessels. UV could also modify thalamic mast cell degranulation, potentially with damaging consequences.

But the mechanisms also imply that no blood-borne photoreceptors or bilirubin (as a photoreceptor, etc.) or mysterious capacity of UVA to reach the retina would be required for UVB or UVA to modify brain function in significant ways (modify the circadian rhythm, thermoregulation, etc.). Changes in the trigeminal system can affect neurotransmission in many parts of the brain, by polysynaptic pathways, and maybe this paper would stimulate some interest in basic research on photobiology or on actual, legitimate treatments for multiple sclerosis, etc. The immunosuppressive effects of UVB are really complex and can go wrong at multiple points along the cascade of events that result from UVB exposure, and the result could be a new type of lupus-like autoimmune disease in some people. It's not a valid immunomodulatory approach. But an implication of the mechanistic analysis might be that drugs modifying CGRP receptor activation could be useful for autoimmune diseases, etc. CGRP is also a very potent vasodilator and produces NOS-independent vasodilation, for example.

In any case, I don't like anti-intellectualism in the context of basic research, and I stand by the validity of my mechanistic discussion in this paper, from an intellectual standpoint. Maybe if more people were not so terrified of discussing some of these topics in a rigorous and objective manner, it would be possible to persuade more people of various public-health-type messages. Instead, the discussion degenerates into proclamations, etc. Again, this paper is not meant to suggest any rationale for doing anything. Maybe people with multiple sclerosis or the like will be more careful about sun exposure and realize that being aware of all the mechanisms can be useful, as far as protecting oneself from the brain damage that could conceivably result from sun exposure, particularly in someone in any kind of disease state.



A Review of the Effects of Cutaneous Exposure to Ultraviolet Radiation On Primary Afferent and Dorsal Horn Neurons: Mechanisms and Effects On Immune Function and Pain
　
Abstract

Background—Following the exposure of the skin to either ultraviolet B (UVB) or ultraviolet A (UVA) radiation, the immunological responses to cutaneously-administered protein antigens or small-molecule haptens can be suppressed systemically. UVB is known to cause sensory C-fibers to release the neuropeptides -calcitonin gene-related peptide (CGRP) and substance P (SP) into the skin, and the release of these neuropeptides contributes to UVB-induced systemic immunosuppression and UVB-induced increases in neurogenic blood flow (i.e. erythema). More specifically, CGRP released from the peripheral terminals of C-fibers, in response to UVB, acts on antigen-presenting cells (APC) migrating from or infiltrating into the skin and contributes to the UVB-induced production of interleukin-10 (IL-10) by these APC. Cutaneous UVB has also been shown to increase the CGRP and SP content in the dorsal horn (DH) of the spinal cord, and this apparent release of CGRP and SP has been suggested to mediate UVB-induced sunburn pain and hyperalgesia. The immunological consequences of CGRP released in the spinal cord has never been investigated but may be relevant for understanding the etiology of multiple sclerosis, which UVB exposure may protect against but also, conceivably, worsen the course of.
Key Conclusions—Researchers have found that ultraviolet radiation (UVR) can influence the electrophysiological activities of and phenotypic expression of proteins by neurons in deeper spinal cord or brain sites (i.e. those that do not receive direct synaptic inputs from C-type, primary afferent neurons). CGRP released in the spinal cord may induce immunosuppressive cytokine production by, or reduce the antigen-presenting/costimulatory capacity of, astrocytes, microglia, or dendritic cells in the CNS. UVR induces low-frequency spontaneous, asynchronous activity in C-fibers innervating the exposed skin and also causes DH neurons to exhibit increases in spontaneous activity. This spontaneous activity releases glutamate, CGRP, and SP in the DH and contributes to UVB-induced primary and secondary hyperalgesia. These neurogenic effects may also contribute to the UVB-induced suppression of pruritus. UVB produces biphasic increases in blood flow in the UV-irradiated skin, and researchers have previously proposed that only the second peak of blood flow is neurogenic (C-fiber-mediated). Numerous pieces of evidence argue against this conclusion and suggest that UVR produces two largely neurogenic phases of increased blood flow. The first peak of blood flow is likely to be both neurogenic and non-neurogenic but is unlikely to be purely non-neurogenic. UVB has been shown to reactivate the herpes simplex virus in the trigeminal ganglia and dorsal root ganglia (DRG), and this reactivation follows a bimodal timecourse. It is likely that UVR first depletes CGRP and SP from C-fibers in the skin and subsequently induces adaptive changes in the cell bodies of DRG neurons and DH neurons, which replenish CGRP and SP stores and produce the second neurogenic phase of erythema via axon reflexes and dorsal root reflexes. UVR has also been found to produce rewarding and pain-reducing effects on the CNS. These effects have generally been attributed to UVR-induced increases in plasma opioid peptides, but it is more likely that these effects occur via purely neurogenic pathways involving the DRG, DH, and supraspinal sites (including the striatum, amygdala, etc.). UVB-induced increases in hormonal vitamin D3 production may modify the C-fiber-mediated effects of UVB and UVA. Hormonal vitamin D3 may, in part, protect against MS by increasing NGF, GDNF, and the low-affinity neurotrophin receptor (p75NTR) in the CNS. A hormonal vitamin D3 analog has been shown to increase the CGRP content in the DRG in an NGF-dependent manner, and the actions of hormonal vitamin D3 may reduce UVB-induced hyperalgesia.

Introduction

Some of the immunosuppressive effects of ultraviolet B radiation (UVB) are thought to be protective against multiple sclerosis (MS) and other autoimmune diseases (McMichael and Hall, 1997). MS is an inflammatory, demyelinating disease of the CNS that is thought to result from an autoimmune response against components of myelin (Storch et al., 1998; Lucchinetti et al., 1996; Lassman et al., 2004). The disease process in MS is also characterized by the degeneration of axons (Trapp et al., 1998) and the loss of neurons (Bozzali et al., 2002; Owens, 2003), and the extent of disability, in people with MS, is closely and positively associated with the loss of axons (Lassman et al., 2004). The demyelination in MS is believed to be largely mediated by CNS-infiltrating T-helper (CD4+) lymphocytes with a pro-inflammatory, Th1, phenotype (Lassman et al., 2004), and UVB reliably suppresses Th1-cytokine production by CD4+ T-cells (Shreedhar et al., 1998) and antigen-presenting cells (APC) (Garssen et al., 1999; Toichi et al., 2002) in regional lymph nodes. The apparent UVB-induced protection against MS has mainly been explained in terms of the immunomodulatory actions of 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] (Hayes, 2000), the hormonally active form of vitamin D3, and, to a lesser extent, -melanocyte stimulating hormone (-MSH) (Friedman, 2004). Increases in the concentrations of both -MSH and 1,25(OH)2D3 have been found to be induced, in the skin, in response to UVB exposure (Lehman et al., 2003; Funasaka et al., 2001).

The implicit assumption has been that any UVB-induced immunosuppressive effects within the CNS must be preceded by, and must also be a consequence of, immunosuppressive effects in regional lymph nodes (RLNs) that drain the UVB-irradiated skin. The UVB-induced elevation in plasma 25-hydroxyvitamin D3 [25(OH)D3] could, as noted by others, clearly produce immunological changes that would originate in the CNS (McCarty, 2006). Microglia, astrocytes, and other cell types can convert 25(OH)D3 into 1,25(OH)2D3 and respond to the newly-produced 1,25(OH)2D3 (Garcion et al., 2002), which can act in a paracrine or autocrine fashion (Garcion et al., 2002). Many, if not most, of the immunosuppressive effects of UVB are known, however, to result from the actions of mediators other than vitamin D3, mediators that include cis-urocanic acid (cis-UCA) (Holán et al., 1998; Sleijffers et al. , 2003) and numerous cytokines (Boonstra et al., 2000), or from changes that overlap in some ways with 1,25(OH)2D3-associated signalling (Lehmann et al., 2004). Although some researchers have discussed the relevance to multiple sclerosis of UVB-induced immunosuppression that is independent of 1,25(OH)2D3 (Chaudhuri, 2005; McMichael and Hall, 1997; Van der Mei et al., 2001), these discussions have frequently been narrowly focused on the effects of -MSH. The involvement of -MSH in UVB-induced immunosuppression is thought to be of somewhat secondary importance and has not been as clearly established as the involvement of other mediators (Shimizu and Streilein, 1994). Many of these immunosuppressive effects, effects that include the UVB-induced expansion of regulatory T-cell populations (Schwarz et al., 2004), have not been discussed, in detailed terms, in the context of multiple sclerosis. UVR has been shown to either prevent damage from EAE (Hauser et al., 1984) or worsen the outcome (Tsunoda et al., 2005), but the complex interactions of UVR with CNS immune privilege and EAE have not been adequately explored (Hauser et al., 1984; Tsunoda et al., 2005).

The induction of DNA damage by UVB is, for example, one effect that is immunosuppressive (Garssen et al., 2000) and that could not be induced by vitamin D3 alone. The development of UVB-induced systemic immunosuppression requires, as one component of a cascade of changes, the formation of pyrimidine dimers in the skin and the migration of DNA-damaged cells to RLNs (Garssen et al., 2000). The UVB-induced formation of cyclopyrimidine dimers in KCs is, furthermore, involved in the UVB-induced increases in IL-10 output by KCs (Nishigori et al., 1996). The enhancement of DNA repair, in UV-irradiated skin, is also known to decrease the systemic immunosuppressive effects of UVB (Garssen et al., 2000). It is notable that 1,25(OH)2D3 and its analogs increase, in various cell types, the protein content of p21CIP1/KIP1/WAF1 (Gumireddy et al., 2003), a protein that induces cell cycle arrest and allows for DNA repair to occur (Weinberg et al., 2002). In the context of DNA damage, 1,25(OH)2D3 could therefore be expected to actually lessen the immunosuppressive effects of UVB.
Apart from the effects of 1,25(OH)2D3 in the CNS (Garcion et al., 2002), UVB-induced immunosuppression has also been investigated only in lymphoid organs that drain tissues outside the brain. The suppressor cells that appear in the spleen, following UVB-irradiation of the skin (Schwarz et al., 2004), may be partially the result of APC migrating from the CNS, and some of these suppressor cells may have been specific to CNS-associated antigens. This possibility has never been evaluated. The UVB-induced expansion of various populations of regulatory T-cells (Schwarz et al., 2004), which have the potential to induce antigen-nonspecific, IL-10-dependent, bystander suppression of other T-cells (Schwarz et al., 2004), and altered APC (Dumas et al., 2000) in RLNs could indeed, as proposed by others (Sharpe, 1986), gradually reinforce tolerance to myelin-associated antigens. These "skin-to-body-to-brain" changes could occur by the mechanisms previously described (Dumas et al., 2000; Sharpe, 1986) and are likely to contribute to UVB-induced neuroimmunomodulation. An example of this type of effect was the amelioration of lupus-associated neuropsychiatric symptoms, and the normalization of the uptake of 18F-2-fluoro-2-deoxyglucose in specific brain regions, by UVA-1 phototherapy in a patient with lupus (Menon et al., 2003). These changes were accompanied by improvements in the clinical lupus scores (Menon et al., 2003). Other studies have found UVA-1-induced reductions in systemic autoantibody titres in patients with lupus (Polderman et al., 2004), reductions that have sometimes been accompanied by improvements in cognitive functioning (McGrath Jr., 2005). Thus, it is reasonable to think that the reductions in autoantibody production occurred primarily in the blood, as a result of UVA-1-induced immunosuppressive changes in the skin, and produced, for example, secondary decreases in endothelial cell activation in cerebral blood vessels. It is also understandable that researchers have traditionally examined UVB-induced changes in T-cells and APC, in the spleen or the lymph nodes draining the skin, with reference to the skin. Given that the skin is used as the initiation and elicitation site for the examination of these protein antigen-specific or hapten-specific immune responses, the focus on skin-derived APC and skin-associated immunity is entirely appropriate. However, researchers have found that UVB can increase the expression of immediate early genes (Gillardon, Wiesner, and Zimmermann, 1992) and the concentrations of neuropeptides in the dorsal horn (DH) of the spinal cord (Gillardon, Schrock, and Morano, 1992). UVR can also influence the mRNA and protein contents of neuropeptides and immediate-early gene products in primary afferent neurons, both in the cell bodies in the dorsal root ganglia (DRG) (Gillardon et al., 1991) and the peripheral branches that innervate the UV-irradiated skin (Benrath et al., 1995; Eschenfelder et al., 1995). The changes in the DH, which occur through UVR-induced actions on DRG sensory C-fibers, could cause immunosuppressive changes to proceed in a "skin-to-brain-to-body" direction. Specifically, the UVB-induced increase in the content of the neuropeptide CGRP in the DH could contribute to the apparent protection against MS.

UVB-induced systemic immunosuppression is known to depend on the neurogenic release of the neuropeptide -calcitonin gene-related peptide (CGRP, or CGRP) into the skin (Kitazawa et al., 2000; Garssen et al., 1998; Hart et al., 2002; Khalil et al., 2001; Niizeki et al., 1997; Legat et al., 2004), but the UVB-induced changes in the CGRP content in the DH and DRG have primarily been viewed in the context of UVB-induced hyperalgesia (Gillardon et al., 1991; Gillardon, Wiesner, and Zimmermann, 1992; Gillardon, Schrock, and Morano, 1992). CGRP that is released from the epidermal or dermal terminals of C-type afferent nociceptive fibers is known to exert, by direct and indirect mechanisms, immunosuppressive effects on APC in the skin (Kitazawa et al., 2000; Niizeki et al., 1997). However, the immunosuppressive implications of the UVB-induced increases in CGRP levels in the spinal cord remain unexplored.
In the context of MS, it is noteworthy that the increased CGRP concentrations, following UVB, in the vicinity of both the peripheral terminals (in the skin) and central terminals (in the DH) of DRG neurons may be partially dependent on NGF. UVB has been shown to increase the expression and release of NGF by keratinocytes (Gillardon et al., 1995) and modify the abundance of NGF receptors on keratinocytes (Bull et al., 1998) and epidermal nerve terminals (Bull et al., 1998). Researchers have also hypothesized that the increased retrograde axonal transport of NGF, from the skin to the cell bodies of DRG neurons, contributes to the UVB-induced increases in CGRP levels in the DH (Gillardon et al., 1995). Although this may be the case, other mechanisms could also be important. The release of NGF from KC, perhaps in concert with the UVB-induced changes in the p75NTR content on the peripheral branches of DRG neurons (Bull et al., 1998), could, by increasing the firing rate of DRG neurons, augment CGRP release, from the central terminals of DRG neurons, or produce phenotypic changes in the cell bodies of DRG neurons. UVR is known to induce C-fibers (Andreev et al., 1994; Eschenfelder et al., 1995; Szolcsányi, 1987) and DH neurons (Urban et al., 1993; Chapman and Dickenson, 1994) to fire, spontaneously, at low-frequencies, and the orthodromic action potentials in C-type, DRG axons are therefore the most obvious "cause" of the UVB-induced release of CGRP from the central terminals of those DRG neurons. The broader question, which will be explored in this paper, is which factors, induced by UVB in the skin, are responsible for the spontaneous action potentials and which factors may help replenish the stores of CGRP, at the central and peripheral terminals of DRG neurons, that are depleted by the UVB-induced, spontaneous activity. KC-derived NGF could exert these effects without undergoing axonal transport to the DRG. NGF, induced in the skin by inflammatory stimuli other than UVR, is known to contribute to spontaneous activity in C-fibers (Djouhri et al., 2001). In addition, exogenous NGF can exert rapid effects on C-fibers, such as the sensitization of C-fibers to capsaicin, that could contribute to hyperalgesia and spontaneous activity (Mendell et al., 2002). Some indirect evidence supports the notion that NGF participates in the UVB-induced increases in C-fiber activity (Khalil et al., 2001), but a UVB-induced increase in the axonal transport of NGF is more hypothetical.

The damage induced by experimental autoimmune encephalomyelitis (EAE) has been found to be less severe after the intracerebroventricular (i.c.v.) injection of exogenous NGF (Triaca et al., 2005) and more severe in rats either autoimmunized against NGF or injected with an anti-NGF antibody. In EAE, an NGF antibody augmented the infiltration of the CNS by pro-inflammatory cells of the immune system (Micera et al., 2000) and appeared to act on progenitor cells participating in the repair of damaged areas (Triaca et al., 2005). The expression and release of CGRP by DRG neurons and B-cells is increased by NGF (Bracci-Laudiero et al., 2002), and many of the anti-inflammatory effects of NGF may be secondary to this CGRP (Bracci-Laudiero et al., 2002). From the standpoint of UVR, the ongoing release of CGRP in the dorsal horn may be accompanied by the release of BDNF. This is because BDNF is known to be co-released with CGRP and SP from the central terminals, in the DH, of C-type, DRG neurons (Malcangio et al., 2003). CGRP has been shown to decrease the severity of experimental autoimmune diabetes (EAD) (Sun et al., 2003) and experimental autoimmune retinitis (Kezuka et al., 2004) and to exert numerous immunosuppressive effects on dendritic cells (Carucci et al., 2000), monocytes (Fox et al., 1997), and T-cells (Boudard et al., 1991). Vitamin D3 could also contribute to the UVB-induced increases in NGF and CGRP. 1,25(OH)2D3, the hormonal form of vitamin D3, and its analogs have been shown to increase the expression of NGF and the abundance of the NGF protein in various cell types, including DRG neurons (Riaz et al., 1999). In addition, the VDR is expressed in lamina I-II of the spinal cord (Stumpf et al., 1988). Thus, 1,25(OH)2D3, produced locally in the skin, or in an autocrine or paracrine fashion in the peripheral or central nervous system, may contribute to the UVR-induced effects on DRG and DH neurons.
This paper will discuss the mechanisms and immunological implications of the UVB-induced increases in the CGRP content of the spinal cord. Specifically, the increased storage and release of CGRP could produce immunosuppressive effects, in the CNS, that would be protective against multiple sclerosis. The UVB-induced increases in CGRP could decrease the release of pro-inflammatory cytokines by Th1 cells, infiltrating the spinal cord, or suppress the antigen-presenting capacity of microglia and infiltrating monocytes. In the DH, UVB-induced release of neuropeptides may also induce the migration, from the CNS, of dendritic cells that are deficient, either because of immaturity and premature migration or because of phenotypic changes induced by CGRP, in their costimulatory capacity. It is also noteworthy that UVR has been shown to exert antihyperalgesic effects in the context of chronic pain (Kaur et al., 2005b) and to produce reinforcing or reward-associated effects on humans (Gambichler et al., 2002a; Feldman et al., 2004; Kaur, Liguori, and Fleischer et al., 2006; Kaur, Liguori, and Lang et al., 2006; Zeller et al., 2006). The supposed rewarding effect of UVR has, however, never been discussed in the context of the known effects of UVR on DRG and DH neurons. There is evidence that UVB can induce the release of -MSH, from the pituitary, via a kind of neurogenic cascade involving the activation of TG neurons, fibers that innervate the cornea and comprise part of the ophthalmic branch of the trigeminal nerve, ciliary ganglion (CG) neurons, and hypothalamic neurons (Hiramoto et al., 2003). In view of these findings, it is possible that the spino-trigemino-parabrachio-amygdaloid and spinohypothalamic pathways may be involved in both the hyperalgesic and antihyperalgesic effects of UVR. The relevance of these possibilities to immune deviation (i.e. immune privilege), across multiple brain regions, will also be briefly discussed. Finally,1,25(OH)2D3, produced locally in the skin or in the DRG and spinal cord from 25(OH)D3 in the systemic circulation, may modify the effects of UVR on CGRP and on the spinal cord. The oral administration of CB1093, a vitamin D analogue, has been found to increase, in an NGF-dependent manner, the CGRP content in sciatic nerve segments of rats (Riaz et al., 1999).
Effects of UVB On Cutaneous Sensory Fibers and the Firing Rates of DRG Neurons
Exposure to UVB increases the content of CGRP alone (Seike et al., 2002) or both CGRP and SP (Legat et al., 2002; Legat et al., 2004) in cutaneous sensory nerve fibers and causes the release of these neuropeptides into the skin (Niizeki et al., 1997). The dermis and basal layer of the epidermis are innervated by the peripheral axons of primary afferent neurons (PANs), whose cell bodies are in the dorsal root ganglia (DRG) (Burbach et al., 2001; Schulze et al., 1997; Reilly et al., 1997). Approximately one million nerve fibers innervate the skin (Krogstad, 1999). UVB has been found to increase the percentages of epidermal nerve fibers that are immunoreactive for CGRP (Legat et al., 2004) and to induce the release of CGRP from nerve terminals into the epidermis (Legat et al., 2004) and the dermis (Niizeki et al., 1997). Researchers have found increases in the release of CGRP from epidermal and dermal afferent fibers, as indicated by an elevated concentration in the skin, that appear as soon as two hours after a single exposure to UVB (Gillardon et al., 1995) and persist for up to seven days (Legat et al., 2004) after the end of a course of multiple exposures. Single exposures to UVB have been shown to increase the content of CGRP in cutaneous fibers at 24 hours (Seike et al., 2002) post-irradiation. Similarly, twelve exposures to UVB over the course of a month roughly tripled the percentage of epidermal fibers containing CGRP (Legat et al., 2004). This peak in CGRP in response to chronic UVB occurred 24 hours after the final UVB treatment (Legat et al., 2004). Although Gillardon et al. (1995) found that UVB transiently reduced the concentration of CGRP in the skin of rats, this decrease was attributed to the UVB-induced release of CGRP from cutaneous nerve fibers (Gillardon et al., 1995). The timecourse of this decrease in CGRP, which emerged as soon as two hours post-irradiation (Gillardon et al., 1995) and began to rebound after 24 hours, has been interpreted as a transient depletion of stored CGRP from epidermal sensory nerve terminals (Gillardon et al., 1995). In the ears of rats exposed to UVB for four weeks, with three low-dose exposures per week, Legat et al. (2002) found data consistent with an adaptive increase in the storage of CGRP per sensory fiber in the epidermis and dermis. Seike et al. (2002) also found an increase in CGRP content within nerve fibers innervating the upper dermis and epidermis at 24 hours after single UVB exposures. These increases became proportionally more pronounced, increasing in a dose-dependent fashion, as the dose of UVB was increased from 0.3 to 0.5 J/cm2, although the increase in the CGRP content appeared to plateau as the dose was increased from 0.5 to 0.7 J/cm2 (Seike et al., 2002). Legat et al. (2002) suggested that the long-term increases in CGRP storage and release, which have been shown to appear after 24 hours and may persist for seven days after the final UVB exposure, are due to the anterograde transport of newly-expressed CGRP from the DRG.

The baseline firing rates and firing thresholds of both thinly-myelinated A-type fibers and unmyelinated C-fibers are sensitive to UV exposure (Andreev et al., 1994) and are likely to be most directly involved in the UVR-induced changes in CGRP content and release. Eschenfelder et al. (1995) found that exposure to a combination of UVB and UVA caused over 35 percent of high-threshold mechanoreceptive C-fibers in the saphenous nerve to exhibit a low-frequency (0.8-1.25 Hz), spontaneous firing pattern. This type of firing pattern, which can occur in the context of UVR exposure and other models of peripheral inflammation, is spontaneous in the sense that endogenous, physiological factors or conditions have become capable of inducing a receptor potential and, thereby, eliciting an orthodromic action potential. The firing of C-fibers at frequencies between 0.1 and 1 Hz, for example, does not produce a sensation of pain (Lynn and Shakhanbeh, 1988; Gybels et al., 1979), may produce a sensation of itch (Torebjörk, 1974), and does produce vasodilation (Lynn and Shakhanbeh, 1988). This spontaneous activity had begun at 24 hours postirradiation, had peaked after 72 hours, and was still slightly increased, above baseline, after 96 hours (Eschenfelder et al., 1995). In rabbits whose shaved ears were exposed to UVR, Szolcsányi (1987) found that polymodal nociceptive C-fibers innervating the irradiated skin of the ear developed a low-frequency pattern of activity. This background activity, at 6.64 impulses per minute or an arithmetic mean of roughly 0.1 Hz, was measurable within five hours post-irradiation in the vast majority of the C-fibers analyzed, although one of the fibers had begun to exhibit activity as soon as 30 minutes after UVR (Szolcsányi, 1987). When bradykinin was administered into the greater auricular artery of UVR-pretreated rabbits, Szolcsányi (1987) found an increase, compared with nonirradiated controls, in the total number of impulses and the "duration" of bradykinin-induced spontaneous activity in polymodal nociceptive C-fibers. These results indicate that UVR can rapidly produce both spontaneous depolarization of C-fibers and sensitization to a given concentration of a substance, such as bradykinin, that is known to induce depolarization. These concepts will subsequently be discussed in more detail. All of the fibers analyzed by Szolcsányi (1987) were fibers of the greater auricular nerve, whose cell bodies are in the cervical DRG in humans. Andreev et al. (1994) found, similarly, that UVR exposure to the rat hindpaw caused C- and A-type fibers of the saphenous nerve to exhibit low-frequency (6-108 discharges per minute, or an arithmetic mean of 0.1-1.8 Hz), spontaneous activity. This ongoing activity, measured at five days post-irradiation, was reduced by the application of morphine and other opioid receptor ligands (Andreev et al., 1994). As discussed below, other anti-hyperalgesic drugs have been shown to influence the UVR-induced sensitization of nociceptive and mechanoreceptive fibers.

Effects of UVB on Neurons in the DRG and DH

In addition to producing changes in sensory fibers in the skin, UVB has been shown to influence the expression of CGRP in the cell bodies of neurons in the DRG (Gillardon et al., 1991) and the content of CGRP in the DH of the spinal cord (Gillardon et al., 1992). Hindpaw exposure to UVB has been shown to decrease the expression of CGRP mRNA in the L3 and L4 DRG (Gillardon et al., 1991) and increase the protein content of CGRP in the medial portion of the superficial dorsal horn (i.e. laminae II-IV), in the L4-L5 lumbar segments (Gillardon et al., 1992). These changes, which were maximal at roughly the same times at which the UVB-induced erythemal skin responses were maximal, were measured in the DRG at 48 hours post-UVB (Gillardon et al., 1991) and, in the DH, from 24 to 96 hours post-UVB (Gillardon et al., 1992). The concentration of CGRP in the DH had increased to 150-160 percent of controls at the first measurement, taken at 24 hours post-UVB, and was still somewhat elevated at both 48 and 96 hours post-UVB (Gillardon et al., 1992). Compared with non-irradiated rats, the UVB-irradiated rats showed both an increase in CGRP and a relative decrement in a higher-molecular-weight CGRP "precursor" (roughly 14.4 kDa) (Gillardon et al., 1992). The increase in CGRP was therefore suggested to have been, in part, a result of the UVB-induced proteolysis of CGRP precursor peptides (Gillardon et al., 1992). The findings may also have resulted from a relative increase in the translation of CGRP, given that CGRP peptides are encoded by mRNA splice variants of mRNA transcripts of the calcitonin (CT) gene.

The effects of low-dose, daily UVB exposure on CGRP release, together with the suppressive effects of UVB on sensory fiber CGRP content in people with psoriasis, are consistent with a longer-term, adaptive decrease in neurogenic inflammation. Legat et al. (2002) found that four weeks of low-dose UVB exposures, given at three times per week, increased the content of CGRP per nerve fiber. However, as noted by the authors, this increase was not accompanied by ongoing, persistent inflammation and edema in the exposed skin (Legat et al. 2002). The authors noted that UVB may, after repeated exposures, decrease the release of CGRP from nerve fibers innervating the exposed skin (Legat et al., 2002). In patients with different subtypes of psoriasis or eczema, UVB decreased the numbers of nerve fibers containing CGRP and also decreased the overall density of nerve fibers (i.e. including those that did not contain CGRP) (Wallengren and Sundler, 2004). Given that the remaining nerve fibers were thicker, Wallengren and Sundler (2004) suggested that UVB had remodeled, rather than produced degeneration in, the sensory innervation of the epidermis and dermis. Wallengren and Sundler (2004) noted that histamine, and the mast cells releasing it, can activate sensory fibers but can also lead to desensitization and neuropeptide depletion in sensory fibers. It was also found that UVB reduced itch and inflammation in the skin of patients (Wallengren and Sundler, 2004), a finding that is consistent with the known antipruritic effects of UVB (Gilchrest et al., 1979; Lim et al., 1997; Holme and Mills, 2001; Kaptanoglu and Oskay, 2003). The suppression of itch was attributed to the UVB-induced changes in cutaneous sensory fibers (Wallengren and Sundler, 2004). Although UVB has been shown to inhibit both the weal and flare responses that are produced by mast cell-derived histamine (Fjellner and Hägermark, 1982), Wallengren and Sundler (2004) noted that UVB often suppresses pruritus in people who have not responded to antihistamines. Consistent with this assessment, Holme and Mills (2001) found that UVB reduced pruritus in a woman whose pruritis had not responded to antihistamines. Interestingly, the woman had also responded to, but had not been able to tolerate, transcutaneous electrical nerve stimulation (Holme and Mills, 2001). Kaptanoglu and Oskay (2003) also found UVB to be effective as an antipruritic in a person who was no longer responding to antihistamines. Together with other findings, which will be discussed in a subsequent section, it should become clear, as noted by Wallengren and Sundler (2004), that the UVB-induced suppression of the weal and flare, components of the so-called axon reflex, cannot be easily attributed to a process such as histamine tachyphylaxis (Wallengren and Sundler, 2004). The relative "histamine-independence" of the UVB-induced antipruritic effects may, for example, result from the central suppression of itch (i.e. in the spinal cord). When UVB is administered to only part of the body surface, the suppression of pruritus is known to be "systemic" and generalized accross the entire body surface (i.e. extending to unexposed sites) (Gilchrest et al., 1979).

UVR has also been shown to influence the concentration of substance P (SP) and the activation of one of its receptors, the neurokinin-1 receptor (NK1R), in the dorsal horn (Polgár et al., 1998; and Thompson et al., 1994). When the spinal cords of rats were removed one day after the unilateral exposure of the rats' (right) hindpaws to UVA, Polgár et al. (1998) found that the substance P content was reduced bilaterally in the L4-L5 lumbar segments and was increased, mainly in the contralateral spinal cord, within the T6-T8 thoracic segments. The distribution of SP in the irradiated rats' spinal cords was compared with the distribution found in non-irradiated controls, rather than the distribution that would have been found prior to irradiation in the experimental group (Polgár et al., 1998). The roughly 50 percent decreases in immunoreactive SP in the lumbar DH were similar on both the ipsilateral side, which contained the central branches of DRG neurons innervating and providing afferent inputs from the irradiated skin, and contralateral side and were found in laminae I and II, in deeper laminae of the DH, and in the lateral spinal nucleus (LSN) (Polgár et al., 1998). In the thoracic segments, the increased SP was more pronounced on the contralateral side than on the ipsilateral side and was most striking, both contralaterally and ipsilaterally, in laminae II and III (Polgár et al., 1998). In view of these results, Polgár et al. (1998) suggested that UVA had increased the production of SP in both the lumbar and thoracic segments of the DH but had, additionally, increased the release of SP in only certain areas. The bilateral decrease of SP in the L4-L5 lumbar DH was viewed as evidence of a UVR-induced increase in SP production and release (Polgár et al., 1998), presumably by L4-L5 DRG neurons that received afferent inputs from the ipsilateral hindpaw and entered the ipsilateral dorsal horn at the L4-L5 segments.
It should be noted that the decrease in SP content in the LSN, in the lumbar spinal cord (Polgár et al., 1998), suggests that UVR can modify the activities of neurons in supraspinal sites, such as the periaqueductal gray matter (PAG). Assuming the decrease in SP content in the LSN was due to an increase in SP release from neurons in the LSN, as proposed by Polgár et al. (1998), the release of SP could have occurred from either descending or ascending pathways (Jiang et al., 1999). Electrophysiological studies suggest that LSN neurons, specifically those that project to supraspinal sites and deliver afferent APs to those supraspinal sites, do not receive direct synaptic inputs from DRG fibers entering the spinal cord (Jiang et al., 1999). The ascending LSN neurons are nonetheless activated, in a polysynaptic manner via intervening, DH interneurons, by stimulation of dorsal root fibers (Jiang et al., 1999). LSN neurons project directly to, and form synapses with, neurons in the PAG (Harmann et al., 1988), thalamus (Battaglia and Rustioni, 1992), hypothalamus (Burstein et al., 1987), and amygdala (Burstein and Potrebic, 1993). The afferent activities of these LSN neurons are, in turn, modified, in the lumbar spinal cord, by mediators released from descending axons of neurons in the raphe nuclei and the PAG (Carlton et al., 1985; Masson et al., 1991). An example of a supraspinal pathway that may be activated by UVR is shown in Fig. 4.

The UVR-induced upregulation of the neurokinin-1 receptor (NK1-R) responsiveness of DH neurons (Thompson et al., 1994) is also consistent with an acute increase in SP-mediated effects. The UVR-induced augmentation of NMDA-R responsiveness in the DH (Thompson et al., 1994; ) also suggests that SP release is acutely increased in response to UVR, given that SP is co-released into the DH, from PAFs, with glutamate (Millan, 1999, Section 10.3). The UVR-induced release of SP in the DH could, in fact, be expected to simultaneously produce an acute decrease in SP content, as found by Polgár et al. (1998), and an increase in the responsiveness of DH neurons to NK1-R and NMDA-R activation. The NK1-R is rapidly internalized in response to SP binding, and the NK1-R is also known to also be redistributed to the plasma membrane after the degradation of SP (Millan, 1999, Section 10.3.2.2). The activation of NK1-Rs and NK2-Rs is nonetheless thought to be more important for the initiation of central sensitization than for its prolongation (Millan, 1999, Section 10.3.2.2). The activation of NK1-Rs by SP produces slow depolarization of DH neurons but can increase the fast and more sustained depolarization induced by NMDA-R activation (Millan, 1999, Section 10.3.2.2; Boxall et al., 1998b). Thus, the release of SP by UVR could account for the findings that UVR augments NMDA-R-mediated activation of DH neurons (Thompson et al., 1994; Thompson et al., 1995; Boxall et al., 1998b). For example, the binding of SP to the NK1-R can, by activating the PLC-IP3-DAG cascade, indirectly enhance the increase in intracellular calcium ([Ca2+]i) that is produced by NMDA-R activation (Millan, 1999, Section 10.3.2.2). Additionally, Boxall et al. (1998b) found evidence that NMDA-R activation, induced in the L5 DH by hindpaw UVR, can exert a "primary" role in sensitizing DH neurons to signals that produce slow depolarization. Hindpaw UVR augmented the depolarization of L5 spinal neurons that had been induced by the administration, intrathecally, of an mGluR1/mGluR5 agonist (Boxall et al., 1998b). This increased responsiveness was largely abrogated by the concurrent administration of an NMDA-R antagonist (Boxall et al., 1998b). The activation of group I mGluRs, such as by the mixed mGluR1/mGluR5 agonist that was used in UVR-treated rodents, generally produces the same slow depolarization of DH neurons and gradual elevation of [Ca2+]i that NK1R activation produces, thereby augmenting the sensitization of DH neurons and contributing to hyperalgesia (Boxall et al., 1998b; Millan, 1999, Section 10.3.2.3). Given that hindpaw UVR enhanced the responsiveness of DH neurons to mGluR1/mGluR5 activation at higher doses of the agonist but did not change the EC50 responses to the agonist, Boxall et al., (1998b) suggested that UVR had probably not upregulated the total numbers of binding sites on the DH neurons. The results of the study implied that hindpaw UVR can, as suggested by the authors (Boxall et al., 1998b), increase NMDA-R-mediated transmission in the spinal cord and thereby increase the mGluR1/mGluR5 responsiveness of DH neurons. More specifically, the authors noted that mGluR1/mGluR5 activation could serve to maintain the activation of NMDA-Rs by phosphorylating the receptors and augmenting protein kinase C (PKC) activation, in much the same way as NK1 receptor activation can augment NMDA-R responses in a PKC-dependent fashion (Boxall et al., 1998b). In contrast, agonists at mGluR3, the mRNA of which was increased by UVA in the DH (Boxall et al., 1998), have been shown to produce antinociceptive effects (Millan, 1999, Section 10.3.2.3). UVR might therefore induce both nociceptive and antinociceptive responses in the spinal cord, and the nociception that occurs through slow depolarization, as in response to CGRP or SP or mGluR receptor activation, is likely to interact with fast (i.e. ionotropic), NMDA/kainate/AMPA-R-mediated, glutamatergic transmission.
It is more difficult, however, to definitively account for the UVA-induced changes in SP content on the contralateral side of the DH. Polgár et al. (1998) implicitly suggested that the UVB-induced activation of primary afferent fibers entering and terminating in the ipsilateral lumbar spinal cord could have induced the depletion of SP from primary afferent fibers terminating in, or at least passing through, the contralateral L4-L5 DH. This is plausible and has been referred to as "volume transmission," whereby mediators are released locally but exert actions distant from the site at which the mediators have been released from (Millan, 1999, Sections 4.7 and 10.3.2.3). The contralateral changes could also have been mediated by the excitation or disinhibition of commissural interneurons, which are abundant in the spinal cord (Sugimoto et al., 1990). Neurons receiving inputs from ipsilateral DRG fibers in the superficial laminae of the DH have been shown to cross the dorsal commissure and influence the contralateral DH in a nearly symmetrical manner (Koltzenberg et al., 1999). Thus, in response to UVR, SP released ipsilaterally in the L4-L5 DH would not have had to diffuse across the midline and induce SP release from primary afferent terminals in the contralateral L4-L5 DH.

Other researchers have found that unilateral exposures to UVR can induce bilateral changes in spinal neurons or peripheral, nociceptive fibers. Thompson et al. (1994) found bilateral thermal and mechanical hyperalgesia on the hindpaws of rats that had been given unilateral, hindpaw UVA. The thermal and mechanical sensitivities were less pronounced on the contralateral hindpaws than the ipsilateral paws, but the timecourses for the changes were similar on both hindpaws (Thompson et al., 1994). Boxall et al. (1998) also found that unilateral UVA exposure to the rat hindpaw produced mechanical hyperalgesia and allodynia on both hindpaws. The peaks of hyperalgesia and allodynia in both hindpaws, measured at 24 hours postirradiation, occurred at roughly the same time that bilateral increases in the mGluR3 mRNA content were found in the lumbar dorsal horn (Boxall et al., 1998). At 24 hours post-UVA, the increase in mGluR3 mRNA was highest in laminae II-IV and lamina I of the L5 lumbar segment but was also found in laminae IV-VII (Boxall et al., 1998). The increases were restricted to laminae I-IV by 48 hours postirradiation, when the sensitivities to mechanical stimuli had begun to normalize (Boxall et al., 1998). Similarly, Gillardon et al. (1992) found that UVB exposed unilaterally to rats' hindpaws increased the concentration of junD mRNA in both the ipsilateral and contralateral sides of the lumbar spinal cord. The junD mRNA content at six hours post-UVB was increased to roughly eight times the level found in non-irradiated rats, an increase that was more or less coincident with the neurogenic vasodilation, or flare, and plasma extravasation that UVB had induced in the irradiated skin (Gillardon et al., 1992).

UVR-induced changes in B1 bradykinin receptor (B1-R) responsiveness have also been found in association with hyperalgesia on the contralateral (non-irradiated) hindpaws of rats (Perkins & Kelly, 1993b). Perkins & Kelly (1993b) found that, compared to controls, unilateral UVA increased the thermal hyperalgesia induced in both the ipsilateral and contralateral hindpaws by an intravenously-injected B1 bradykinin receptor (B1-R) agonist. In the absence of treatment with the B1-R agonist, thermal hyperalgesia was only significant on the UV-irradiated (ipsilateral) hindpaw and was still present, following its peak at 48 hours post-UVA, at 96 hours post-UVA (Perkins & Kelly, 1993b). Gougat et al. (2004) found that the thermal hyperalgesia induced by unilateral hindpaw UVA, which could be reduced by up to 85 percent by a small-molecule B1-R antagonist, was only significant on the irradiated side at 48 hours post-UVA. Although the same half-duration dose of UVA (6,210 mJ/cm2) used by both groups of investigators might account for the absence (Gougat et al., 2004) or "subclinical" character (Perkins & Kelly, 1993b) of the observed contralateral hyperalgesia, Perkins & Kelly (1993b) found that the hyperalgesia measured on the contralateral hindpaw was highest at 24 hours post-UVA and had decreased by 48 hours post-UVA. Given this earlier disappearance of BK-R hyperresponsiveness on the contralateral side, contralateral hyperalgesia may have developed in the animals studied by Gougat et al. (2004) and subsided by the 48 hour time point at which the B1-R antagonist was administered.

UVR has also been shown to induce spontaneous activity in, and increases in the excitability of, DH neurons (Urban et al., 1993; Chapman and Dickenson, 1994). In general, the spontaneous activity in DH neurons is induced by C-fiber activity but soon becomes independent of changes in C-fiber activity. In other words, the timecourse of the increases in spontaneous C-fiber activity, induced by UVR exposure, should not be assumed to parallel the increases in the firing rates of DH neurons. Szolcsányi (1987) measured spontaneous APs in PMN C-fibers that began at 30 minutes post-UVR and were well-developed across the interval of 2.5-5 hours post-UVR, but the timecourse of C-fiber activity has not been analyzed across the entire, up-to-7-day timecourse of UVR-induced hyperalgesia and blood flow increases. Eschenfelder et al. (1995) began analyzing the spontaneous APs in C-fibers, which were found to occur at 0.8-1.25 Hz, at 24 hours post-UVB and took daily measurements on each of the four subsequent days. The percentage of C-fibers showing spontaneous activity was highest at 72 hours post-UVB and declined almost to baseline by 5 days post-UVB, but the changes in the activities of C-fibers were not monitored over the first 24 hours following UVB (Eschenfelder et al., 1995). Urban et al. (1993) found that the spontaneous activity of WDR neurons, in the DH, was largely independent of C-fiber inputs at 5-7 days post-UVR, as indicated by the nonsignificant effect of dorsal rhizotomy on WDR neuron activity, but was even somewhat independent, albeit nonsignificantly, over the 1-3 day, post-UVB interval. Thus, the activities of DH neurons do not simply parallel, temporally, the activities of C-fibers. There is also some evidence that the UVR-induced changes in the spontaneous firing rates of DH neurons does not correlate, and may even vary inversely, with the excitability of DH neurons. For example, Chapman and Dickenson (1994) found that the UV-induced augmentation of one form of C-fiber wind-up, which involves measuring the excitability of DH neurons and is thought to reflect central sensitization, increased between 3 and 5 days post-UVR but was accompanied by a nonsignificant decrease in the mean frequency, from 2.5 to 1.86 Hz, of the spontaneous APs in DH neurons in the L1-L3 segments. Similarly, Chapman and Dickenson (1994) measured UVR-induced electrophysiological changes, in the DH, that were consistent with allodynia and that were disconnected from artificially-induced changes in C-fiber activity. Consistent with allodynia, Chapman and Dickenson found decreased thresholds in A fibers innervating the UV-irradiated hindpaw and increased numbers of action potentials induced in WDR neurons in response to a fixed-duration, three-times-threshold stimulation of A fibers. However, the firing of WDR neurons was not significantly augmented, compared to non-irradiated controls, in response to peripheral stimulation of C-fibers (Chapman and Dickenson, 1994). These results indicate that, in the context of central sensitization, the firing rates of DH neurons may increase or decrease in ways that do not reliably correlate with changes in spontaneous C-fiber activity.

From a practical standpoint, it should also be evident that the time post-UVR cannot be used to predict the degree to which UVR-induced hyperalgesia is peripherally-mediated or centrally-mediated. Thompson et al. (1994) found that hindpaw UVR produced an augmentation of A-wind-up, an electrophysiological change that is consistent with allodynia, in the hemisected spinal cords of rats at 24 hours post-UVR. Nociceptive responses that are consistent with central sensitization can therefore be established, in response to UVR exposure, rather quickly. Other results, apart from the effects of UVR, shed light on the capacity for C-fiber activity to rapidly induce central sensitization. Klede et al. (2003) found that 1 Hz stimulation of high-threshold, mechanically-insensitive C-fibers produced punctante SMHA and allodynia that were both centrally-mediated, but only the allodynia was clearly dependent on ongoing C-fiber stimulation. Although the results of Chapman and Dickenson (1994) imply that UVR-induced allodynia can become partially independent of C-fiber activity, the emergence of both allodynia and punctate SMHA after only 30 minutes of C-fiber stimulation (Klede et al., 2003) highlights the rapidity with which central sensitization can emerge and become, particularly in the case of punctate SMHA, partially independent of C-fiber activity.

Bradykinin and Early, UVB-Induced Action Potentials in C-fibers

Other evidence suggests that BK exerts direct effects on nociceptive PAFs, and these effects may contribute to the early induction by UVR of spontaneous activity in C-fibers. In the UVB-irradiated skin of humans, Eisenbarth et al. (2004) found that the neurogenic, axon reflex-associated vasodilation was enhanced, compared to controls, in response to the localized perfusion of B1-R and B2-R agonists into the irradiated skin. While the B1-R agonist-induced, subjective pain ratings were also enhanced at the perfusion site, the vasodilation induced by the BK-R agonists at the infusion site was not augmented by UVB (Eisenbarth et al., 2004). By inserting microdialysis catheters intracutaneously (i.e. intradermally) into the irradiated skin, Eisenbarth et al. (2004) were able to assess BK-R-induced vasodilation at both the site of BK-R agonist perfusion and the skin surrounding the perfusion site. Both sites of analysis were within the boundary of the irradiated skin, and the experiments were performed at 24 hours post-UVB (Eisenbarth et al., 2004). UVB evidently induced an increase in the B1-R or B2-R responsiveness of C-fibers or augmented the release of other C-fiber-activating mediators from KC within the perfusion site. The absence of local vasodilation largely excludes a UVB-induced increase in the responsiveness of ETC or SMC to the direct actions of BK. If UVB had induced BK-R hyperresponsiveness in both C-fibers and ETC or SMC, the BK-R agonists would be expected to have produced non-neurogenic vasodilation at the perfusion site and neurogenically-mediated vasodilation in the skin at which the flare was induced. It is noteworthy that vasodilation both within and surrounding the irradiated skin could be explained by axon reflexes induced by UVB, and Eisenbarth et al. (2004) was, therefore, evaluating the capacity of BK-R agonists to augment UVB-induced axon reflexes within the irradiated site (see Fig. 2). The effects of BK-R agonists that were unique to the UVB-irradiated subjects, and that were not found in non-irradiated controls, can be seen as "UVB-specific."

These and other results suggest that bradykinin may contribute to the spontaneous C-fiber activity induced by UVB. In rabbits whose ears had been exposed to UVR, Szolcsányi (1987) found that the administration of BK into the greater auricular artery produced a greater number and duration of spontaneous action potentials, compared to nonirradiated controls, in polymodal nociceptive C-fibers of the greater auricular nerve. In non-irradiated rabbits, bradykinin injected intra-arterially also induced "spontaneous" activity polymodal nociceptive C-fibers (Szolcsányi, 1987). In the context of the spontaneous APs that were measured in C- and A-fibers of the saphenous nerve at 5 days post-UVR, Andreev et al. (1994) noted that BK and other early mediators were unlikely to contribute directly to the activation of PAFs at such a late time point. As discussed below, however, the early activation of C-fibers by BK and other mediators may be necessary for C-fiber activity to be sustained by UVR-induced, late-phase mediators. For example, authors have proposed that NGF may sustain the release of CGRP and SP from C-fibers (Khalil et al., 2002) or be responsible for the ongoing APs in C-fibers at 5 days post-UVR (Andreev et al., 1994).

Biphasic Increases in Blood Flow: Are The Effects of Prostaglandins Non-neurogenic or Just Non-Activating?

The difficulty arises in attempting to reconcile the rapidly-induced release of CGRP and SP in the irradiated site (Benrath et al., 1995) with the apparent capsaicin-insensitivity of the first 24 hours of UVB-induced erythema. Some investigators have found that UVB induces two phases of increased blood flow within the irradiated site (Benrath et al., 1995; Benrath et al., 2001), and the early and late peaks of erythema were attributed, respectively, to non-neurogenic and neurogenic mediators (Benrath et al., 2001). The early peak increase in blood flow occurred at 1 hour post-UVB in rats (Benrath et al., 1995) and 12 hours post-UVB in humans (Benrath et al., 2001), and the second peak occurred in rats at 24 hours (Benrath et al., 1995) and in humans at 36 hours (Benrath et al., 2001) post-UVB. When human skin was treated with topical capsaicin for four days and was exposed to UVB on the day after the final capsaicin treatment, there was no reduction in blood flow in the irradiated skin until 24 hours post-UVB (Benrath et al., 2001). In part because the first peak of erythema was insensitive to capsaicin pretreatment, which depletes SP and CGRP from C-fiber terminals, it was suggested that primarily the second phase of erythema was neurogenic (Benrath et al., 2001). This neurogenic SP and CGRP release was proposed to be mediated by axon reflexes (Benrath et al., 2001). Given that the UVB-induced increases in HA and prostaglandins have been found to decrease to pre-UVB concentrations within 18-24 hours post-UVB and that the administration of COX inhibitors have been shown to only inhibit erythema within the first 24-36 hours post-UVB, the early phase of erythema was attributed to the non-neurogenic, direct vasodilatory actions of PGs and HA (Benrath et al., 2001).

Although the early phase of UVB-induced erythema is likely to be partially non-neurogenic in origin, there is considerable evidence that neurogenic effects of UVB begin almost immediately after exposure. For example, the increase in blood flow to human skin that was up to 10 mm outside the irradiated border, a neurogenically-mediated effect, began, in the absence of capsaicin pretreatment, at 9 hours post-UVB (Benrath et al., 2001). This is a more telling result than the limited attenuation of the early erythema by capsaicin, in part because the effects of topical capsaicin are less predictable than intradermal capsaicin and are less reliable in humans than in rodents (Szallasi and Blumberg, 1999). For example, Munn et al. (1997) found that topical capsaicin did not alter SP immunoreactivity in the skin of humans. In contrast, intradermal capsaicin was found to produce a pronounced decrease in SP immunoreactivity and degeneration of SP-containing nerve fibers (Szallasi and Blumberg, 1999). Szallasi and Blumberg (1999) also noted that human skin is between 4 and 8 times less permeable to topical capsaicin than rat skin. Benrath et al. (2001) also noted that the topical capsaicin preparation that was used had previously been shown to be too low in potency to completely deplete the stores of neuropeptides from sensory fibers. It is conceivable that capsaicin partially depleted the SP and CGRP stores from C-fibers and that the UVB-induced increases in BK, PGE2, TNF-, IL-1, and other early mediators depleted the remaining stores, thereby explaining the monophasic increase in blood flow, as found by Benrath et al. (2001), in capsaicin-pretreated, UVB-exposed human skin.

The finding that capsaicin pretreatment did not attenuate the UVB-induced thermal hyperalgesia within the first 24 hours post-UVB (Benrath et al., 2001) could also be interpreted as evidence of an early neurogenic effect of UVB. At first glance, this might appear to be consistent with the view that predominantly non-neurogenic mechanisms occur during the first 24 hours. Although a nonsignificant attenuation of the UVB-induced thermal hyperalgesia (THA) was apparent only after a 24 hour delay in capsaicin pretreated skin, the UVB-induced THA increased monophasically over the first 24 hours in capsaicin-pretreated skin and did not begin after a 24 hour latent period (Benrath et al., 2001). An antihyperalgesic effect of capsaicin pretreatment began to emerge after the 24-hour point (Benrath et al., 2001), and this suggests that a minimal desensitization of C-fiber responses occurred in response to capsaicin. In other words, the capsaicin pretreatment may have blunted the CGRP release produced by early, neurogenically-acting mediators. As the CGRP and SP stores were being replenished over roughly the first 24 hours post-UVB, after having been partially depleted by topical capsaicin pretreatment, the newly-replenished CGRP and SP could have produced the monophasically-emerging hyperalgesic and vasodilatory effects found by Benrath et al. (2001). This could have caused the first 24 hours of UVB-induced changes to appear "non-neurogenic," when in fact the C-fibers would have been "running on empty" and have been acted upon by BK and other early mediators.

Although the UVB-induced increases in the concentrations of PGs and HA do return to baseline levels within 24-36 hours, their early effects on C-fibers may contribute to hyperalgesia and neurogenic vasodilation at later time points. Benrath et al. (2001) suggested that UVB-induced HA, PGs and other mediators may have contributed to THA and MHA, by sensitizing C-fibers, during the first 36 hours post-irradiation but that substance P and other C-fiber-derived neuromediators may have sustained the THA after that point. Similarly, Eschenfelder et al. (1995) suggested that HA and PGs induced transiently by UVB were likely to primarily influence the erythemal and edematous responses during the time period, namely the first 8-24 hours, their concentrations were elevated. UVB-induced PGE2 could conceivably have reduced the thresholds, in C-fibers, required for C-fiber-activating stimuli (i.e. heat) or substances to induce action potentials, and this PG-mediated effect could conceivably produce THA without inducing AR-mediated vasodilation. Prostaglandins are known to sensitize C-fibers to subthreshold depolarization without, themselves, inducing APs (Millan, 1999). It is known that COX inhibitors administered more than 24-36 hours post-UVB are ineffective in reducing the erythemal response, but COX inhibitors administered immediately after UVB produced relatively stable anti-erythemal effects over 48 hours (Eschenfelder et al., 1995). It was noted that UVB-induced PGE2 and PGF2 levels may remain elevated up to 48 hours post-irradiation (Eschenfelder et al., 1995), and this could support the suggestion that a mixture of neurogenic and non-neurogenic factors contribute to UVB-induced blood flow, in animals, at the 24-hour time point (Eschenfelder et al., 1995). But given that the anti-erythemal effect of early, COX-inhibitor treatment persisted past the 36 hour point, which corresponded to the more robustly neurogenic, late blood flow peak in humans (Benrath et al., 2001), it is possible that the early effects of PGs on C-fibers facilitate the later, sustained phase of neurogenic inflammation and vasodilation. PGs and HA were not suggested to have been the exclusive or obligatory mediators of the early vasodilatory and hyperalgesic effects (Benrath et al., 2001), and PGs and HA are not the only mediators that are rapidly induced by UVB and that could act on C-fibers. BK, for example, is induced by UVB in a matter of minutes (Kang-Rotondo et al., 1996) and is known to be able to induce action potentials, and not simply induce sensitization, in C-fibers (Banik et al., 2001). Andreev et al. (1994) suggested that the pro-inflammatory cytokines and NGF could have produced the spontaneous activities of C- and A-fibers that they had measured at five days post-UVB. Additionally, the neurogenic vasodilation that was evident in humans at 9 hours post-UVB (Benrath et al., 2001) is a reliable sign that C-fibers were being depolarized, not merely sensitized, and that APs were being induced in them.

Following UVB+UVA exposure to the rat hindpaw, for example, the CGRP content of the skin decreased as soon as 2 hours post-UVR (Gillardon et al., 1995). The decrease in CGRP was nearly maximal by 6 hours post-UVB, was maximal by 12 hours post-UVB, and was starting to increase at 24 hours post-UVB (Gillardon et al., 1995). The authors suggested that UVB had depleted CGRP from the skin by inducing the release of CGRP from cutaneous nerve fibers (Gillardon et al., 1995). Although the early peak in blood flow occurred in rats at 1 hour post-UVB, the early increase had not declined, to a between-peak minimum, until 12 hours post-UVB (Benrath et al., 1995). In the same experiments, the s.c. (systemic) administration of the CGRP receptor antagonist CGRP(8-37), the NOS inhibitor L-NAME, or a combination of the two agents reduced UVB-induced blood flow as soon as 1 hour post-irradiation. The injection of the NK1-R antagonist, CP-96,345, at 1 hour post-UVB also reduced blood flow. Similarly, Eschenfelder et al. (1995) found that each of two SP receptor antagonists, administered separately and intradermally to UVA+UVB-exposed ears, decreased the resulting edema as soon as 6 hours post-irradiation and the erythemal response by 12 hours post-UVB.

When the bimodal qualities of UVR-induced herpesvirus reactivation are viewed in the context of the above controversy, UVR can be seen as inducing two phases of neurogenic changes. In experimental reactivation of HSV by UVB, for example, there is frequently a biphasic pattern to the reactivation and the appearance of lesions (Bernstein et al., 1997; Spruance and Kriesel, 2002; Burkhart and Burkhart, 2005). The first peak occurs at roughly 24 hours post-UVB and has been attributed to HSV reactivation in the skin (Burkhart and Burkhart, 2005). The second peak requires about four days, post-UVB, to occur in humans, and this has been attributed to the time required for anterograde axonal transport, as from the DRG or TG to the skin, of HSV proteins (Burkhart and Burkhart, 2005). The second neurogenic phase induced by UVB may, also in the context of CGRP release, be an "axonal transport phase," in which stores of CGRP and other proteins are transported from the cell bodies of DRG neurons to the peripheral terminals, in the skin via anterograde axonal transport, and the central terminals of DRG neurons. This second phase would replenish the depleted stores of CGRP and allow for other phenotypic changes to occur on DRG neurons.
　
Central Suppression of Itch by Pain, and Suppression of Both Itch and Pain by UVR

Given that UVB has been shown to suppress pathological itch (pruritus) and induce hyperalgesia, at least acutely, it is possible to view some of the effects of UVR in terms of the inverse relationship between itch and pain (Ikoma et al., 2003). Eisenbarth et al. (2004) found that activation of BK receptors, with the use of BK receptor agonists, did not produce itch sensation in UVB irradiated skin but did produce itch in nonirradiated controls. UVB irradiation did produce sensitization to the neurogenic effects, both in terms of the axon reflex flare and the hyperalgesia, of the BK agonists, and the authors suggested that UVB-induced hyperalgesia may have produced centrally-mediated suppression of itch (Eisenbarth et al., 2004). The generalized, inverse relationship between itch and pain is known to be tied to opiodergic effects (Ikoma et al., 2003), but it is necessary to specify the site at which the change in opiodergic signalling is occcurring. This is currently a difficult task in the context of the effects of UVR, given that the effects of UVR interact with peripheral opiodergic signalling but are also likely to affect spinally and supraspinally mediated opiodergic signalling.

Exposure to UVR has been shown to increase the expression of the proopiomelanocortin (POMC) gene and the production of -endorphin and -lipotropin by KC (Wintzen et al., 1996), and UVR also modifies the responsiveness of PAFs to opioid receptor ligands (Andreev et al., 1994). The interpretation of these findings has, however, been a source of significant confusion. In early research, exposures of large areas of skin to UVR were found to increase the plasma concentrations of -endorphin and other POMC-derived peptides (Levins et al., 1983). Given the large areas of UVR-exposed skin, the UVR-induced release of the peptides from KCs was thought to be extensive enough to increase the systemic concentrations of the peptides. Researchers have not consistently found changes in the concentrations of plasma opioid peptides in UVR-exposed humans (Wintzen et al., 2001; Gambichler et al., 2002b), and a number of relevant issues have not been addressed adequately. Some researchers have found reinforcing or reward-like effects of UVR and have explained the results in terms of an opioidergic effect (Kaur, Liguori, and Fleischer et al., 2006; Kaur, Liguori, and Lang et al., 2006). To the extent that an increase in the plasma concentration of -endorphin or another opioidergic peptide could contribute to these subjective effects, the opioid peptides would have to cross the blood-brain barrier and ultimately exert a generalized augmentation of -opioidergic activity in one or another supraspinal sites. An increase in plasma -endorphin would, most simply, not be a reliable indication of supraspinally-mediated antinociception. Apart from this issue, the increases in plasma opioids and POMC-derived mediators are more likely to be mediated by UVR-induced effects on the spinohypothalamic tract or on other neuronal populations that influence pituitary function. For example, the UVB-induced increase in plasma -MSH, following UVB exposure to the eyes alone, was found to be blocked by hypophysectomy or ciliary gangliectomy (Hiramoto et al., 2003). In addition, both the antipruritic effects of UVB and the effects of UVR on pain thresholds or C-fiber activity are broadly consistent with, at least in the short term, the antagonism of -opioidergic activity.

It is noteworthy that the notion of UVB as rewarding stimulus (Kaur, Liguori, and Fleischer et al., 2006; Kaur, Liguori, and Lang et al., 2006) is generally consistent with the augmentation of anti-nociceptive pathways and is clearly inconsistent with an escalation of centrally-mediated hyperalgesia. Researchers have found evidence that UVR can suppress pain for several hours after exposure (Kaur et al., 2005b) and that UVR can be used to prevent post-herpetic neuralgia (Jalali et al., 2006). While the investigation of UVR-induced antinociceptive and reward-associated effects is a valid avenue of research, one problem seems to be the assumption that the reward-associated effects must be primarily or exclusively opioidergic. This is not the case. Becerra et al. (2001) noted that ascending nociceptive pathways can themselves activate neurons in the ventral striatum and nucleus accumbens, meaning that nociceptive stimuli activate dopaminergically-mediated reward centers in the brain (Gear et al., 1999; Becerra et al., 2001). These effects could occur via the activation of spinothalamic tract neurons or by the direct activation of striatal neurons, given that neurons in the lateral dorsal horn of primates and rats are known to form direct synaptic connections with striatal neurons (Newman et al., 1996)
Gillardon et al. (1992) suggested, explicitly, that the apparent UVB-induced release of CGRP into the DH could both contribute to UVB-induced hyperalgesia and, implicitly, activate descending, -opioidergic, supraspinally-mediated, pathways. Although researchers have not investigated the involvement of specific supraspinal sites in the hyperalgesic or anti-hyperalgesic effects of UVB, it is likely that chronic treatment with UVB, particularly at high-doses, would activate and produce changes in neurons that exert descending influences on nocisponsive, DH neurons.

Although any UVR-induced changes in opioidergic activity in spinal or supraspinal neurons are poorly understood and are likely to be complex, UVR has produced changes in the responses of C-fibers to opioid-receptor (OR) ligands. Andreev et al. (1994) found that the application of either of two -OR agonists, morphine and DAGOL, or the -OR agonist, U-69593, to the peripheral terminals of C-fibers and A-fibers, in UV-irradiated skin, reduced the frequencies of spontaneous APs in the fibers. These reductions, measured at five days post-UVR, were naloxone-reversible (Andreev et al., 1994). These findings are relevant to a discussion of the supposed naltrexone-sensitivity of the addictive or reinforcing effects of UVR. Given that the peripheral, hyperalgesic effects of UVR are naloxone-sensitive, one would expect naloxone to disinhibit and essentially "unblind" the peripheral component of UVR-induced hyperalgesia. The naloxone would clearly be expected to amplify, both at the peripheral and spinal level, the hyperalgesic effects of ongoing UVR. Given that the behavior could be modified by merely the peripheral actions of systemically-administered naloxone, it is inappropriate to conclude that UVR produces some sort of mechanistically-nonspecific, opioidergically- and supraspinally-mediated reinforcing effect.

In the context of multiple sclerosis, it is noteworthy that IL-10, in addition to other mediators, are likely to contribute to both immunosuppressive and antihyperalgesic effects of UVR. The UVB-induced production of IL-10, by APC and other cells, is known to depend on UVB-induced CGRP release (Kitazawa et al., 2000). Given that exogenous IL-10 was found to counteract UVB-induced hyperalgesia (Saadé et al., 2000), it follows that UVB-induced endogenous IL-10 may also contribute to the supposed antihyperalgesic effects of chronic UVB. There is already some evidence that UVB can produce cytokine "cascades," in the cell bodies and central branches of trigeminal ganglion (TG) neurons, that parallel the pattern of UVB-induced cytokine production in the skin. Shimeld et al. (1999) found that UVB induced TNF- and IL-6 production, by satellite cells, in the TG of mock-inoculated mice (i.e. those that had not been infected with HSV). This transient inflammatory response is unlikely to persist, given that IL-6 knockout mice are known to have reduced IL-10 production in response to UVB (Nishimura et al., 1999). TNF-, induced in response to UVB-induced CGRP release, is also known to be required, via the TNF--induced migration of LC, for UVB-induced local immunosuppression (Niizeki et al., 1997). In addition, the UVB-induced synthesis of 1,25(OH)2D3 by KCs is known to be dependent on the UVB-induced increases in the TNF-a content in the skin (Lehman et al., 2004). A similar progression of TNF-- and IL-6-induced anti-inflammatory effects may occur in the spinal cord or other sites in the CNS.

Interactions With Vitamin D-Mediated Effects And Prospects For Further Research

Vitamin D may interact in a number of ways with the neurogenic effects of UVR. The oral administration of the vitamin D analogue CB1093, in a rat model of diabetic neuropathy, was found to increase, compared to untreated diabetic rats, the CGRP, substance P, and NGF protein concentrations in segments of the sciatic nerve (Riaz et al., 1999). In non-diabetic rats, compared to non-diabetic rats not treated with CB1093, the oral CB1093 also increased the content of CGRP and NGF in the sciatic nerve fibers, the NGF content in the soleus muscle, and the NGF mRNA in the skin from the hindlimb foot. The increases in CGRP were thought to be NGF-dependent and secondary to the CB1093-induced increase in the NGF protein content in the sciatic nerve fibers (Riaz et al., 1999).

VDR ligands are also known to induce GDNF and the low-affinity neurotrophin receptor (p75NTR) in various cell types found in the CNS, and these changes could modify the effects of UVR on the spinal cord. VDR ligands have been shown to induce the expression of the low-affinity neurotrophin receptor (p75NTR) in glioma cells (Naveilhan et al., 1996a), the expression and protein content of 75NTR in the developing brain (Eyles et al., 2003), and the p75NTR mRNA content of cultured oligodendrocytes and astrocytes (Baas et al., 2000). The p75NTR receptor binds all members of the neurotrophin family and, in concert with TrkA, is thought to be involved in the retrograde axonal transport, at least by L4 and L5 DRG neurons, of NGF (Delcroix et al., 1997). In the developing brains of rats whose mothers were depleted of dietary vitamin D3, the levels of p75NTR mRNA were reduced by 30 percent and the p75NTR protein content, in four separate brain regions, was almost completely depleted (Eyles et al., 2003). Maternal vitamin D3 depletion also reduced the concentration of the free NGF protein by 17 percent and the concentration of free GDNF by 25 percent (Eyles et al., 2003). Although VDR ligands do not appear to regulate the expression or protein content of BDNF, it is noteworthy that p75NTR is thought to be important for the trophic actions of BDNF. The induction of NGF and perhaps other neurotrophins by 1,25(OH)2D3 in the skin may also contribute to the effects of UVB on DRG neurons. Tacalcitol, a 1,25(OH)2D3 analog, has been shown to increase NGF expression the release of NGF by cultured human keratinocytes (Fukuoka et al., 2001). Fukuoka et al. (2001) suggested that VDR ligands may, by increasing neurotrophin expression in the skin, have potential in the treatment of peripheral neuropathy. It is interesting that GDNF has the potential to treat neuropathic pain (Sah et al., 2005), and GDNF has been found to upregulate CGRP expression by sensory neurons without inducing hyperalgesia (Ramer et al., 2003). 1,25(OH)2D3 has been shown to increase GDNF production by numerous cell types (Naveilhan et al., 1996b).

UVR has also been shown to modify the p75NTR content of sensory fibers (Bayerl et al., 1997; Moll et al., 1994), and these effects may or may not be partially dependent on UVB-induced 1,25(OH)2D3. For example, the p75NTR content in cutaneous nerve fibers was found to be reduced at 24 hours post-UVR in humans with UV-induced dermatitis (Bayerl et al., 1997) and also to be reduced in the dermal nerve fibers of normal humans at 48 hours post-UVB (Moll et al., 1994). The induction of NT-3 and NT-4/5 production in KCs exposed to UVB (Marconi et al., 2003), and the induction of NT-3 production in UVA-irradiated KCs (Marconi et al., 2003), are other effects of UVR that are strikingly similar to the effects of 1,25(OH)2D3 on neurotrophin production (Neveu et al., 1994). 1,25(OH)2D3 was found to upregulate NT-3 and NT-4 production by astrocytes (Neveu et al., 1994). Again, the effects of UVB and UVA on NT-3 are probably independent of 1,25(OH)2D3, but this does not preclude an effect of 1,25(OH)2D3 on KC neurotrophin production or on C-fibers in the skin. Given these remarkable similarities between the effects of VDR ligands and the effects of UVR on neurotrophin production, it is not unreasonable to suspect some local effects of UVB-induced 1,25(OH)2D3 on sensory fibers.
Interestingly, Plotnikoff and Quigley (2003) recently found that 93 percent of people who sought medical treatment for nonspecific, musculoskeletal pain were clinically deficient in vitamin D3. This was consistent with previous reports of muscle pain, occurring in conjunction with muscle weakness, in people with vitamin D3 deficiency (Plotnikoff and Quigley, 2003). Although the musculoskeletal pain in vitamin D3 deficiency was suggested to be secondary to the abnormalities in bone structure that are associated with vitamin D3 deficiency, the pain could also be the result of central sensitization and be explained in terms of the UVB-induced changes in DRG and DH neurons.

Conclusions

In summary, the UVR-induced release of CGRP in the spinal cord is likely to contribute to secondary hyperalgesia, to the antihyperalgesic effects of chronic UVR, to the suppression of pruritus by UVB, and to the reward-associated effects of UVR. CGRP released in the DH and in other spinal and supraspinal sites may also induce immunosuppressive and antihyperalgesic changes in astrocytes, microglia, or dendritic cells in the CNS. In the skin, prostaglandins are likely to sensitize C-fibers to the action-potential-inducing effects of histamine and bradykinin. The low-frequency spontaneous activity induced in C-fibers appears to induce central sensitization in DH neurons, and the firing rates of DH neurons appear to become independent of the firing rates of the C-type neurons that provide direct or indirect (i.e. converging, polysynaptic) inputs to the DH neurons. UVR-induced spontaneous activity in C-fibers has been shown to begin by 30 minutes post-irradiation, and this indicates that the timecourse alone cannot be used to distinguish between primary (peripheral) and secondary (peripheral and central) hyperalgesia. A closer examination of the timecourse, the times post-irradiation at which UVR-induced changes in the peripheral and central nervous systems occur, will nonetheless be important for future research examining changes in astrocytes or microglia. The abundance of evidence suggests that the UVR-induced changes in the CNS should be examined at time points as early as 0.5-2 hours post-irradiation and followed, at various time points, for at least five or more days post-irradiation. The immunological changes in the lymph nodes draining the CNS, such as in the posterior cervical triangle, may require considerably more time to appear.

With regard to multiple sclerosis and the generalized immunological effects of centrally-released CGRP, a number of additional conclusions follow from the experimental results discussed in this paper. First, the exposure of UVB or UVA to the eyes is likely to be especially perilous for individuals with multiple sclerosis. Although it is well known that essentially no UVB wavelengths penetrate deeper than the cornea and that little UVA penetrates deeper than the iris and lens (Sliney, 1997), the cornea is densely innervated by, for example, sensory fibers whose cell bodies are in the TG. The classical map of somatosensory "two-point discrimination" can be used as a crude indicator of the potential for direct, neurological damage induced by UVB or UVA exposure, and casual exposure of the face, and particularly the eyes, should probably be aggressively avoided by individuals with MS (i.e. given the potential, via polysynaptic changes induced by UVR through the TG, caudal trigeminal nucleus, and ciliary ganglia-to-Edinger-Westphal nucleus, for brainstem damage or neurovascular events and increases in BBB permeability). The effects of UVR on the CNS are likely to be especially potent following exposure to the eyes and, additionally, especially difficult to research, given that most sun-derived UVR reaches the surface of the eyes as diffuse UVB and UVA (i.e. after Rayleigh and Mie scattering by gases in the atmosphere of the Earth) (Sliney, 1997). Diffuse UVR would be exceptionally difficult to produce by artificial light sources, and the application of direct UVB or UVA from an artificial source would be potentially so damaging, not only to the eyes but also to the CNS, as to be unethical.

Notwithstanding these conclusions, it is possible that CGRP release, either related to or dependent on NGF trafficking from the skin, plays some role in the protection against MS among individuals younger than a certain age. Although UVR is unlikely to be useful in a therapeutic context, the involvement of NGF and CGRP in UVR-induced neuroimmunology could have implications for the use of vitamin D receptor analogs in MS. The NGF-inducing effects of VDR ligands have not previously been viewed as being relevant to the protection against MS, but the NGF-dependent augmentation, by VDR ligands, of CGRP production in the CNS is likely to have therapeutic implications for MS. Given that cutaneous UVR exposure is likely to exert complex, polysynaptic changes in the activities of various neuronal populations throughout the CNS, future research in rodents should focus on the synaptic actions, rather than increases in the plasma concentrations, of -MSH and other POMC-derived peptides.

References

Aldskogius H, Arvidsson J, and Grant G. The reaction of primary sensory neurons to peripheral nerve injury with particular emphasis on transganglionic changes. Brain Res 1985; 357(1): 27-46.
Andreev, N. et al. Opioids suppress spontaneous activity of polymodal nociceptors in rat paw skin induced by ultraviolet irradiation. Neuroscience 1994; 58(4): 793-8.
Anthoney DA, Bone I, and Evans TR. Inflammatory demyelinating polyneuropathy: a complication of immunotherapy in malignant melanoma. Ann Oncol 2000; 11(9): 1197-200.
Arvidson B. Retrograde transport of horseradish peroxidase in sensory and adrenergic neurons following injection into the anterior eye chamber. J Neurocytol 1979; 8(6): 751-64.
Avis SP and Pryse-Phillips WE. Sudden death in multiple sclerosis associated with sun exposure: a report of two cases. Can J Neurol Sci 1995; 22(4): 305-7.
Baas, D. et al. Rat oligodendrocytes express the vitamin D(3) receptor and respond to 1,25-dihydroxyvitamin D(3). Glia 2000; 31: 59-68.
Banik, R.K. et al. B2 receptor-mediated enhanced bradykinin sensitivity of rat cutaneous C-fiber nociceptors during persistent inflammation. J Neurophysiol 2001; 86: 2727-2735.
Battaglia, G., and Rustioni, A. Substance P innervation of the rat and cat thalamus. II. Cells of origin in the spinal cord. J Comp Neurol 1992; 315: 473-486.
Bayerl, C. et al. Immunohistochemical characterization of HSP, alpha-MSH, Merkel cells and neuronal markers in acute UV dermatitis and acute contact dermatitis in vivo. Inflamm Res 1997; 46(10): 409-11.
Becerra, L. et al. Reward circuitry activation by noxious thermal stimuli. Neuron 2001; 32(5): 927-46.
Benrath J, Gillardon F, and Zimmermann M. Differential time courses of skin blood flow and hyperalgesia in the human sunburn reaction following ultraviolet irradiation of the skin. Eur J Pain 2001; 5(2): 155-67.
Benrath J, Zimmermann M, and Gillardon F. Substance P and nitric oxide mediate wound healing of ultraviolet photodamaged rat skin: evidence for an effect of nitric oxide on keratinocyte proliferation. Neurosci Lett 1995; 200(1): 17-20.
Bernstein, D.I. et al. Effect of foscarnet cream on experimental UV radiation-induced herpes labialis. Antimicrobial Agents and Chemotherapy 1997; 41(9): 1961-1964.
Boonstra, A. et al. UVB irradiation modulates systemic immune responses by affecting cytokine production of antigen-presenting cells. International Immunology 2000; 12(11): 1531-1538.
Bothwell, M. Neurotrophin function in skin. J Investig Dermatol Symp Proc 1997; 2(1): 27-30.
Boudard, F. et al. Inhibition of mouse T-cell proliferation by CGRP and VIP: effects of these neuropeptides on IL-2 production and cAMP synthesis. J Neurosci Res 1991; 29: 29.
Boxall, S.J. et al. Enhanced expression of metabotropic glutamate receptor 3 messenger RNA in the rat spinal cord during ultraviolet irradiation induced peripheral inflammation. Neuroscience 1998; 82(2): 591-602.
Boxall, S.J. et al. MGluR activation reveals a tonic NMDA component in inflammatory hyperalgesia. Neuroreport 1998b; 9: 1201-1203.
Bozzali, M. et al. Quantification of brain grey matter damage in different MS phenotypes by use of diffusion tensor MR imaging. AJNR Am J Neuroradiol 2002; 23: 985-988.
Bracci-Laudiero, L. et al. NGF modulates CGRP synthesis in human B-lymphocytes: a possible anti-inflammatory action of NGF? Journal of Neuroimmunology 2002; 123: 58-65.
Bull, H.A. et al. Expression of nerve growth factor receptors in cutaneous inflammation. British Journal of Dermatology 1998; 139: 776-783.
Burbach, G.J. et al. The neurosensory tachykinins substance P and neurokinin A directly induce keratinocyte nerve growth factor. J Invest Dermatol 2001; 117: 1075-1082.
Burkhart, C.N., and Burkhart, C.G. Bimodal temporal distribution of herpetic reactivation. Mayo Clin Proc 2005; 80(2): 287-288.
Burstein, R. et al. Direct somatosensory projections from the spinal cord to the hypothalamus and telencephalon. J Neurosci 1987; 7: 4159-4164.
Burstein, R., and Potrebic, S. Retrograde labeling of neurons in the spinal cord that project directly to the amygdala or the orbital cortex in the rat. J Comp Neurol 1993; 335: 469-485.
Carlton, S.M. et al. Funicular trajectories of brainstem neurons projecting to the lumbar spinal cord in the monkey (Macaca fascicularis): a retrograde labeling study. J Comp Neurol 1985; 241: 382-404.
Carucci, J.A. et al. Calcitonin gene-related peptide decreases expression of HLA-DR and CD86 by human dendritic cells and dampens dendritic cell-driven T cell-proliferative responses via the type I calcitonin gene-related peptide receptor. J Immunol 2000; 164: 3494-3499.
Chan KY, Beck TW, Mullen P, and Dong WK. Increased axonal transport of trigeminal ganglion proteins in an electrode model. Invest Ophthalmol Vis Sci 1989; 30(9): 2050-5.
Chapman, V., and Dickenson, A.H. Enhanced responses of rat dorsal horn neurons after UV irradiation of the hindpaw; roles of the NMDA receptor. Neurosci Lett 1994; 176(1): 41-4.
Chaudhuri, A. Why we should offer routine vitamin D supplementation in pregnancy and childhood to prevent multiple sclerosis. Medical Hypotheses 2005; 64(3): 608-618.
Constantinescu CS. Melanin, melatonin, melanocyte-stimulating hormone, and the susceptibility to autoimmune demyelination: a rationale for light therapy in multiple sclerosis. Med Hypotheses 1995; 45(5): 455-8.
Davis, A.J. et al. The induction of des-Arg9-bradykinin-mediated hyperalgesia in the rat by inflammatory stimuli. Brazilian J Med Biol Res 1994; 27: 1793-1802.
Delcroix, J-D. et al. Diabetes and axotomy-induced deficits in retrograde axonal transport of nerve growth factor correlate with decreased levels of p75LNTR protein in lumbar dorsal root ganglia. Molecular Brain Research 1997; 51: 82-90.
Djouhri, L. et al. Time course and nerve growth factor dependence of inflammation-induced alterations in electrophysiological membrane properties in nociceptive primary afferent neurons. The Journal of Neuroscience 2001; 21(22): 8722-8733.
Drummond PD. Attenuation of axon reflexes to compound 48/80 after repeated iontophoresis of compound 48/80 in skin of the human forearm. Skin Pharmacol Appl Skin Physiol 2003; 16(4): 263-70.
Dumas M and Jauberteau-Marchan MO. The protective role of Langerhans' cells and sunlight in multiple sclerosis. Med Hypotheses 2000; 55(6): 517-20.
Eisenbarth, H. et al. Sensitization to bradykinin B1 and B2 receptor activation in UV-B irradiated human skin. Pain 2004; 110: 197-204.
Eschenfelder CC, Benrath J, Zimmermann M, and Gillardon F. Involvement of substance P in ultraviolet irradiation-induced inflammation in rat skin. Eur J Neurosci 1995; 7(7): 1520-6.
Eyles, D. et al. Vitamin D3 and brain development. Neuroscience 2003; 118(3): 641-53.
Feldman, S.R. et al. Ultraviolet exposure is a reinforcing stimulus in frequent indoor tanners. J Am Acad Dermatol 2004; 51: 45-51.
Ferguson IA, Schweitzer JB, and Johnson EM Jr. Basic fibroblast growth factor: receptor-mediated internalization, metabolism, and anterograde axonal transport in retinal ganglion cells. J Neurosci 1990; 10(7): 2176-89.
Fjellner, B., and Hägermark, Ö. Influence of ultraviolet light on itch and flare reactions in human skin induced by histamine and the histamine liberator compound 48/80. Acta Derm Venereol 1982; 62: 137-140.
Fox, F.E. et al. Calcitonin gene-related peptide inhibits proliferation and antigen presentation by human peripheral blood mononuclear cells: effects on B7, interleukin-10, and interleukin-12. J Invest Dermatol 1997; 108: 43-48.
Friedman, A.P. et al. Do hyporesponsive genetic variants of the melanocortin 1 receptor contribute to the etiology of multiple sclerosis? Medical Hypotheses 2004 62: 49-52.
Fukuoka, M., et al. Tacalcitol, an active vitamin D3, induces nerve growth factor production in human epidermal keratinocytes. Skin Pharmacol Appl Skin Physiol 2001; 14(4):226-33.
Funasaka Y. et al. The effect of thioredoxin on the expression of proopiomelanocortin-derived peptides, the melanocortin-1 receptor and cell survival of normal human keratinocytes. J Invest Derm Symp Proc 2001; 6: 32-37.
Gallar J, Garcia de la Rubia P, Gonzalez GG, and Belmonte C. Irritation of the anterior segment of the eye by ultraviolet radiation: influence of nerve blockade and calcium antagonists. Curr Eye Res 1995; 14(9): 827-35.
Gambichler, T. et al. Impact of UVA exposure on psychological parameters and circulating serotonin and melatonin. BMC Dermatol 2002a; 2:6.
Gambichler, T. et al. Plasma levels of opioid peptides after sunbed exposures. Br J Dermatol 2002b; 147: 1207-11.
Garcion, E. et al. New clues about vitamin D functions in the nervous system. Trends Endocrinol Metab 2002; 13: 100-105.
Garssen J, Buckley TL, and Van Loveren H. A role for neuropeptides in UVB-induced systemic immunosuppression. Photochem Photobiol 1998; 68(2): 205-10.
Garssen, J. et al. UVB exposure-induced systemic modulation of Th1 and Th2 mediated immune responses. Immunology 1999; 97: 506.
Garssen, J. et al. Transcription-coupled and global genome repair differentially influence UV-B-induced acute skin effects and systemic immunosuppression. J Immunol 2000; 164: 6199-6205.
Gear, R.W. et al. Pain-induced analgesia mediated by mesolimbic reward circuits. J Neurosci 1999; 19(16): 7175-81.
Ghadge GD, Roos RP, Kang UJ, Wollmann R, Fishman PS, Kalynych AM, Barr E, and Leiden JM. CNS gene delivery by retrograde transport of recombinant replication-defective adenoviruses. Gene Ther 1995; 2(2): 132-7.
Gilchrest, B.A. et al. Ultraviolet phototherapy of uremic pruritus. Ann Intern Med 1979; 91: 17-21.
Gillardon F, Eschenfelder C, Rush RA, and Zimmerman M. Increase in neuronal Jun immunoreactivity and epidermal NGF levels following UV exposure of rat skin. Neuroreport 1995; 6(9): 1322-4.
Gillardon F, Morano I, and Zimmermann M. Ultraviolet irradiation of the skin attenuates calcitonin gene-related peptide mRNA expression in rat dorsal root ganglion cells. Neurosci Lett 1991; 124(2): 144-7.
Gillardon F, Schrock H, Morano I, and Zimmermann M. Long-term increase in CGRP levels in rat spinal dorsal horn following skin ultraviolet irradiation. A mechanism of sunburn pain? Ann N Y Acad Sci 1992; 657: 493-6.
Gillardon F, Wiesner RJ, and Zimmermann M. Expression of the junD proto-oncogene in the rat spinal cord and skin following noxious cutaneous ultraviolet irradiation. Neurosci Lett 1992; 136(1): 87-90.
Gougat, J. et al. SSR240612 [(2R)-2-[((3R)-3-(1,3-benzodioxol-5-yl)-3-[[(6-methoxy-2-naphthyl)sulfonyl]amino]propanoyl)amino]-3-(4-[[2R,6S)-2,6-dimethylpiperidinyl]methyl]phenyl)-N-isopropyl-N-methylpropanamide hydrochloride], a new nonpeptide antagonist of the bradykinin B1 receptor: biochemical and pharmacological characterization. J Pharmacol Exp Ther 2004; 309(2): 661-9.
Gumireddy, K. et al. Anti-proliferative effects of 20-epi-vitamin-D3 analogue, KH1060 in human neuroblastoma: induction of RAR- and p21Cip1. Cancer Letters 2003; 190: 51-60.
Gybels, J. et al. A comparison between the discharge of human nociceptive nerve fibres and the subject's rating of his sensations. J Physiol (London) 1979; 292: 193-206.
Harbison JW, Calabrese VP, and Edlich RF. A fatal case of sun exposure in a multiple sclerosis patient. J Emerg Med 1989; 7(5): 465-7.
Harmann, P.A. et al. Collaterals of spinothalamic tract cells to the periaqueductal gray: a fluorescent double-labeling study in the rat. Brain Res 1988; 441: 87-97.
Hart PH, Townley SL, Grimbaldeston MA, Khalil Z, and Finlay-Jones JJ. Mast cells, neuropeptides, histamine, and prostaglandins in UV-induced systemic immunosuppression. Methods 2002; 28(1): 79-89.
Hauser, S.L. et al. Prevention of experimental allergic encephalomyelitis (EAE) in the SJL/J mouse by whole body ultraviolet radiation. J Immunol 1984; 132(3): 1276-1281.
Hayes CE, Cantorna MT, and DeLuca HF. Vitamin D and multiple sclerosis. Proc Soc Exp Biol Med 1997; 216(1): 21-7.
Hayes, C.E. Vitamin D: a natural inhibitor of multiple sclerosis. Proc Nutr Soc 2000; 59(4): 531-5.
Hiramoto, K. et al. Ultraviolet B irradiation of the eye activates a nitric oxide-dependent hypothalamopituitary proopiomelanocortin pathway and modulates functions of -melanocyte-stimulating hormone-responsive cells. J Invest Dermatol 2003; 120: 123-127.
Holán, V. et al. Urocanic acid enhances IL-10 production in activated CD4+ T cells. J Immunol 1998; 161: 3237-3241.
Holme, S.A., and Mills, C.M. Crotamiton and narrow-band UVB phototherapy: novel approaches to alleviate pruritus of breast carcinoma skin infiltration. Journal of Pain and Symptom Management 2001; 22(4): 803-805.
Ikoma, A. et al. Neurophysiology of pruritus: interaction of itch and pain. Arch Dermatol 2003; 139: 1475-1478.
Jalali, M.H. et al. Broad-band ultraviolet B phototherapy in zoster patients may reduce the incidence and severity of postherpetic neuralgia. Photodermatol Photoimmunol Photomed 2006; 22(5): 232-7.
Jiang, M.C. et al. Cellular properties of lateral spinal nucleus neurons in the rat L6–S1 spinal cord. J Neurophysiol 1999; 81: 3078-3086.
Kaptanoglu, A.F., and Oskay, T. Ultraviolet B treatment for pruritus in Hodgkin’s lymphoma. Journal of the European Academy of Dermatology and Venereology 2003; 17: 489-490.
Kang-Rotondo, C.H. et al. Upregulation of nitric oxide synthase in cultured human keratinocytes after ultraviolet B and bradykinin. Photodermatol Photoimmunol Photomed 1996; 12(2): 57-65.
Kaur, M. et al. Side effects of naltrexone observed in frequent tanners: Could frequent tanners have ultraviolet-induced high opioid levels? J Am Acad Dermatol 2005a; (May): 916.
Kaur, M. et al. Indoor tanning relieves pain. Photodermatol Photoimmunol Photomed 2005b; 21(5): 278.
Kaur, M.; Liguori, A.; Fleischer, A.B. et al. Plasma -endorphin levels in frequent and infrequent tanners before and after ultraviolet and non-ultraviolet stimuli. J Am Acad Dermatol 2006; 54(5): 919-920.
Kaur, M.; Liguori, A.; Lang, W. et al. Induction of withdrawal-like symptoms in a small randomized, controlled trial of opioid blockade in frequent tanners J Am Acad Dermatol 2006; 54: 709-11.
Kessler JA, Bell WO, and Black IB. Interactions between the sympathetic and sensory innervation of the iris. J Neurosci 1983; 3(6): 1301-7.
Kezuka, T. et al. Peritoneal exudate cells treated with calcitonin gene-related peptide suppress murine experimental autoimmune uveoretinitis via IL-10. J Immunol 2004; 173: 1454-1462.
Khalil Z, Townley SL, Grimbaldeston MA, Finlay-Jones JJ, and Hart PH. Cis-Urocanic acid stimulates neuropeptide release from peripheral sensory nerves. J Invest Dermatol 2001; 117(4): 886-91.
Kitazawa, T. et al. Hapten-specific tolerance promoted by calcitonin gene-related peptide. J Invest Dermatol 2000; 115: 942-948.
Kloos L, Sillevis Smitt P, Ang CW, Kruit W, and Stoter G. Paraneoplastic ophthalmoplegia and subacute motor axonal neuropathy associated with anti-GQ1b antibodies in a patient with malignant melanoma. J Neurol Neurosurg Psychiatry 2003; 74(4): 507-9.
Klyce SD, Beuerman RW, and Crosson CE. Alteration of corneal epithelial ion transport by sympathectomy. Invest Ophthalmol Vis Sci 1985; 26(4): 434-42.
Koltzenburg, M. et al. Does the right side know what the left is doing? Trends Neurosci 1999; 22(3): 122-7.
Krogstad, A.L. Neurogenic control of blood flow and histamine release in psoriatic skin. Acta Derm Venereol (Stockholm) 1999; 79(suppl. 203): 3-43.
Laborde JM, Dando WA, and Teetzen ML. Climate, diffused solar radiation and multiple sclerosis. Soc Sci Med 1988; 27(3): 231-8.
Lassmann, H., and Ransohoff, R.M. The CD4-Th1 model for multiple sclerosis: a critical re-appraisal. Trends Immunol 2004; 25(3): 132-7.
Legat FJ, Griesbacher T, Schicho R, and Althuber P, Schuligoi R, Kerl H, and Wolf P. Repeated subinflammatory ultraviolet B irradiation increases substance P and calcitonin gene-related peptide content and augments mustard oil-induced neurogenic inflammation in the skin of rats. Neurosci Lett 2002; 329(3): 309-13.
Legat, F.J. et al. The role of calcitonin gene-related peptide in cutaneous immunosuppression induced by repeated subinflammatory ultraviolet irradiation exposure. Experimental Dermatology 2004; 13: 242-250.
Lehmann B. et al. Demonstration of UVB-induced synthesis of 1,25-dihydroxyvitamin D3 in human skin by microdialysis. Arch Dermatol Res 2003; 295: 24-28.
Lehman B. et al. Role for tumor necrosis factor- in UVB-induced conversion of 7-dehydrocholesterol to 1,25-dihydroxyvitamin D3 in cultured keratinocytes. Journal of Steroid Biochemistry & Molecular Biology 2004; 89-90: 561-565.
Levins, P.C. et al. Plasma beta-endorphin and beta-lipotropin response to ultraviolet radiation. Lancet 1983; 2: 166.
Lim, H.W. et al. UVB phototherapy is an effective treatment for pruritus in patients infected with HIV. J Am Acad Dermatol 1997; 37: 414-417.
Liu XG, Morton CR, Azkue JJ, Zimmermann M, and Sandkuhler J. Long-term depression of C-fibre-evoked spinal field potentials by stimulation of primary afferent Adelta-fibres in the adult rat. Eur J Neurosci 1998; 10(10): 3069-75.
Livermore A, Hummel T, Pauli E, and Kobal G. Perception of olfactory and intranasal trigeminal stimuli following cutaneous electrical stimulation. Experientia 1993; 49(10): 840-2.
Lucas RM and Ponsonby AL. Ultraviolet radiation and health: friend and foe. Med J Aust 2002; 177(11-12): 594-8.
Lucchinetti, C. et al. Distinct patterns of multiple sclerosis pathology indicates heterogeneity in pathogenesis. Brain Pathol 1996; 6: 259-274.
Lynn, B. and Shakhanbeh, J. Neurogenic in£ammation in the skin of the rabbit. Agents Actions 1988; 25: 228-230.
Malcangio, M. et al. A common thread for pain and memory synapses? Brain-derived neurotrophic factor and trkB receptors. Trends in Pharmacological Sciences 2003; 24(3): 116-121.
Marconi, A. et al. Expression and function of neurotrophins and their receptors in cultured human keratinocytes. J Invest Dermatol 2003; 121: 1515-1521.
Masson, R.L., Jr. et al. Descending projections to the rat sacrocaudal spinal cord. J Comp Neurol 1991; 307: 120-130.
McCarty, M.F. Down-regulation of microglial activation may represent a practical strategy for combating neurodegenerative disorders. Med Hypoth 2006; 67(2): 251-269.
McGrath Jr., H. Elimination of anticardiolipin antibodies and cessation of cognitive decline in a UV-A1-irradiated systemic lupus erythematosus patient. Lupus 2005; 14: 859-861.
McMichael AJ and Hall AJ. Does immunosuppressive ultraviolet radiation explain the latitude gradient for multiple sclerosis? Epidemiology 1997; 8(6): 642-5.
McMichael AJ and Hall AJ. Multiple sclerosis and ultraviolet radiation: time to shed more light. Neuroepidemiology 2001; 20(3): 165-7.
Mendell, L.M. et al. Diversity of neurotrophin action in the postnatal spinal cord. Brain Research Reviews 2002; 40: 230-239.
Menetrey D and Basbaum AI. Spinal and trigeminal projections to the nucleus of the solitary tract: a possible substrate for somatovisceral and viscerovisceral reflex activation. J Comp Neurol 1987; 255(3): 439-50.
Menon, Y. et al. Reversal of brain dysfunction with UV-A1 irradiation in a patient with systemic lupus. Lupus 2003; 12: 479-482.
Micera, A. et al. Nerve growth factor antibody exacerbates neuropathological signs of experimental allergic encephalomyelitis in adult Lewis rats. J Neuroimmunol 2000; 104: 116-123.
Miki K, Fukuoka T, Tokunaga A, and Noguchi K. Calcitonin gene-related peptide increase in the rat spinal dorsal horn and dorsal column nucleus following peripheral nerve injury: up-regulation in a subpopulation of primary afferent sensory neurons. Neuroscience 1998; 82(4): 1243-52.
Millan, M.J. The induction of pain: an integrative review. Progress in Neurobiology 1999; 57: 1-164.
Moll, I. et al. Effects of ultraviolet B radiation on cytoskeletal and adhesion molecules in human epidermis. Photodermatol Photoimmunol Photomed 1994; 10(1): 26-32.
Munn, S.E. et al. The effect of topical capsaicin on substance P immunoreactivity: A clinical trial and immunohistochemical analysis. Acta Derm Venereol (Stockh) 1997; 77: 158-159.
Naveilhan, P. et al. 1,25-Dihydroxyvitamin D3 regulates the expression of the low-affinity neurotrophin receptor. Brain Res Mol Brain Res 1996a; 41: 259-268.
Naveilhan, P. et al. 1,25-Dihydroxyvitamin D3, an inducer of glial cell line-derived neurotrophic factor. Neuroreport 1996b; 7: 2171-2175.
Neveu, I. et al. 1,25-dihydroxyvitamin D3 regulates NT-3, NT-4 but not BDNF mRNA in astrocytes. Neuroreport 1994; 6: 124-126.
Newman, H.M. et al. Direct spinal projections to limbic and striatal areas: anterograde transport studies from the upper cervical spinal cord and the cervical enlargement in squirrel monkey and rat. J Comp Neurol 1996; 365(4): 640-58.
Niizeki, H. et al. Calcitonin gene-related peptide is necessary for ultraviolet B-impaired induction of contact hypersensitivity. J Immunol 1997; 159: 5183-5186.
Niizeki H, Kurimoto I, and Streilein JW. A substance P agonist acts as an adjuvant to promote hapten-specific skin immunity. J Invest Dermatol 1999; 112(4): 437-42.
Nishigori, C. et al. Evidence that DNA damage triggers interleukin-10 cytokine production in UV-irradiated murine keratinocytes. Proc Natl Acad Sci USA 1996; 93: 10354.
Nishimura, N. et al. Defective immune response and severe-skin damage following UVB irradiation in interleukin-6-deficient mice. Immunology 1999; 97: 77.
Oktar BK and Alican I. Modulation of the peripheral and central inflammatory responses by alpha-melanocyte stimulating hormone. Curr Protein Pept Sci 2002; 3(6): 623-8.
Owens, T. The enigma of multiple sclerosis: inflammation and neurodegeneration causes heterogenous dysfunction and damage. Curr Opin Neurol 2003; 16: 259-265.
Perkins, M.N. et al. Antinociceptive activity of the bradykinin B1 and B2 receptor antagonists, des-Arg9, [Leu8]-BK and HOE 140, in two models of persistent hyperalgesia in the rat. Pain 1993a; 53: 191-197.
Perkins, M.N., and Kelly, D. Induction of bradykinin B1 receptors in vivo in a model of ultra-violet irradiation-induced thermal hyperalgesia in the rat. Br J Pharmacol 1993b; 110: 1441-1444.
Pinessi L, Rainero I, de Gennaro T, Violante A, Cassano D, Matera L, and Cesano A. Cerebrospinal fluid and plasma concentrations of POMC-related peptides in multiple sclerosis. Ann N Y Acad Sci 1992; 650: 351-4.
Plotnikoff, G.A., and Quigley, J.M. Prevalence of severe hypovitaminosis D in patients with persistent, nonspecific musculoskeletal pain. Mayo Clin Proc 2003; 78(12): 1463-70.
Polderman, M.C.A. et al. Efficacy of UVA-1 cold light as an adjuvant therapy for systemic lupus erythematosus. Rheumatology 2004; 43: 1402-1404.
Polgár, E. et al. Alterations of substance P immunoreactivity in lumbar and thoracic segments of rat spinal cord in ultraviolet irradiation induced hyperalgesia of the hindpaw. Brain Res 1998; 786(1-2): 248-51.
Ponsonby AL, McMichael A, and van der Mei I. Ultraviolet radiation and autoimmune disease: insights from epidemiological research. Toxicology 2002; 181-182: 71-8.
Ramer, M.S. et al. Glial cell line-derived neurotrophic factor increases calcitonin gene-related peptide immunoreactivity in sensory and motoneurons in vivo. Eur J Neurosci 2003; 18(10): 2713-21.
Reilly, D.M. et al. The epidermal nerve fibre network. characterization of nerve fibres in human skin by confocal microscopy and assessment of racial variations. Br J Dermatol 1997; 137: 163-170.
Reynier-Rebuffel AM, Mathiau P, Callebert J, Dimitriadou V, Farjaudon N, Kacem K, Launay JM, Seylaz J, and Abineau P. Substance P, calcitonin gene-related peptide, and capsaicin release serotonin from cerebrovascular mast cells. Am J Physiol 1994; 267(5 Pt 2): R1421-9.
Riaz, S. et al. A vitamin D3 derivative (CB1093) induces nerve growth factor and prevents neurotrophic deficits in streptozotocin-diabetic rats. Diabetologia 1999; 42: 1308-1313.
Roelcke U, Hornstein C, Hund E, Schmitt HP, Siess R, Kaltenmaier M, Fassler J, and Meinck HM. "Sunbath polyneuritis": subacute axonal neuropathy in perazine-treated patients after intense sun exposure. Muscle Nerve 1996; 19(4): 438-41.
Saadé, N.E. et al. Modulation of ultraviolet-induced hyperalgesia and cytokine upregulation by interleukins 10 and 13. British Journal of Pharmacology 2000; 131: 1317-1324.
Sah, D.W. et al. New approaches for the treatment of pain: the GDNF family of neurotrophic growth factors. Curr Top Med Chem 2005; 5(6): 577-83.
Sandyk R and Awerbuch GI. Nocturnal plasma melatonin and alpha-melanocyte stimulating hormone levels during exacerbation of multiple sclerosis. Int J Neurosci 1992; 67(1-4): 173-86.
Scholzen TE, Brzoska T, Kalden DH, O'Reilly F, Armstrong CA, Luger TA, and Ansel JC. Effect of ultraviolet light on the release of neuropeptides and neuroendocrine hormones in the skin: mediators of photodermatitis and cutaneous inflammation. J Investig Dermatol Symp Proc 1999; 4(1): 55-60.
Schulze, E. et al. Immunohistochemical detection of human skin nerve fibers. Acta Histochem 1997; 99: 301-309.
Schwarz, A. et al. Ultraviolet radiation-induced regulatory T-cells not only inhibit the induction but can suppress the effector phase of contact hypersensitivity. J Immunol 2004; 172: 1036-1043.
Seiffert K and Granstein RD. Neuropeptides and neuroendocrine hormones in ultraviolet radiation-induced immunosuppression. Methods 2002; 28(1): 97-103.
Seike, M. et al. Increased synthesis of calcitonin gene-related peptide stimulates keratinocyte proliferation in murine UVB-irradiated skin. Journal of Dermatological Science 2002; 28: 135-143.
Sharpe RJ. The low incidence of multiple sclerosis in areas near the equator may be due to ultraviolet light induced suppressor cells to melanocyte antigens. Med Hypotheses 1986; 19(4): 319-23.
Shimeld, C. et al. Reactivation of herpes simplex virus type 1 in the mouse trigeminal ganglion: an in vivo study of virus antigen and cytokines. Journal of Virology 1999; 73(3): 1767-1773.
Shimizu, T., and Streilein, J.W. Influence of alpha-melanocyte stimulating hormone on induction of contact hypersensitivity and tolerance. J Dermatol Sci 1994; 8(3):187-93.
Shreedhar, V.K. et al. Origin and characteristics of ultraviolet-B radiation-induced suppressor T-lymphocytes. J Immunol 1998; 161: 1327-1335.
Sleijffers, A. et al. Epidermal cis-urocanic acid levels correlate with lower specific cellular immune responses after Hepatitis B vaccination of ultraviolet B-exposed humans. Photochemistry and Photobiology 2003; 77(3): 271-275.
Sliney, D.H. Ultraviolet radiation effects upon the eye: problems of dosimetry. Radiation Protection Dosimetry 1997; 72(3): 197-206.
Slominski A, Pisarchik A, Zbytek B, Tobin DJ, Kauser S, and Wortsman J. Functional activity of serotoninergic and melatoninergic systems expressed in the skin. J Cell Physiol 2003; 196(1): 144-53.
Spruance, S.L., and Kriesel, J.D. Treatment of herpes simplex labialis. Herpes 2002; 9(3): 64-69.
Storch, M.K. et al. Multiple sclerosis: in situ evidence for antibody and complement mediated demyelination. Ann Neurol 1998; 43: 465-471.
Streilein JW, Okamoto S, Sano Y, and Taylor AW. Neural control of ocular immune privilege. Ann N Y Acad Sci 2000; 917: 297-306.
Stumpf, W.E. et al. 1,25(OH)2 vitamin D3 sites of action in spinal cord and sensory ganglion. Anat Embryol (Berl) 1988; 177(4): 307-10.
Sun, W. et al. Intramuscular transfer of naked calcitonin gene-related peptide gene prevents autoimmune diabetes induced by multiple low-dose streptozotocin in C57BL mice. Eur J Immunol 2003; 33: 233-242.
Suzuki N, Hardebo JE, and Owman C. Trigeminal fibre collaterals storing substance P and calcitonin gene-related peptide associate with ganglion cells containing choline acetyltransferase and vasoactive intestinal polypeptide in the sphenopalatine ganglion of the rat. An axon reflex modulating parasympathetic ganglionic activity? Neuroscience 1989; 30(3): 595-604.
Szallasi, A., and Blumberg, P.M. Vanilloid (capsaicin) receptors and mechanisms. Pharmacological Reviews 1999; 51(2): 159-211.
Thompson, S.W. et al. Injury-induced plasticity of spinal reflex activity: NK1 neurokinin receptor activation and enhanced A- and C-fiber mediated responses in the rat spinal cord in vitro. J Neurosci 1994; 14(6): 3672-87.
Thompson, S.W. et al. No evidence for contribution of nitric oxide to spinal reflex activity in the rat spinal cord in vitro. Neurosci Lett 1995; 188(2): 121-4.
Toichi E. et al. Cell surface and cytokine phenotypes of skin immunocompetent cells involved in ultraviolet-induced immunosuppression. Methods 2002; 28: 104-110.
Torebjörk, H.E. Afferent C units responding to mechanical, thermal and chemical stimuli in human nonglabrous skin. Acta Physiol Scand 1974; 92: 374-390.
Townley SL, Grimbaldeston MA, Ferguson I, Rush RA, Zhang SH, Zhou XF, Conner JM, Finlay-Jones JJ, and Hart PH. Nerve growth factor, neuropeptides, and mast cells in ultraviolet-B-induced systemic suppression of contact hypersensitivity responses in mice. J Invest Dermatol 2002; 118(3): 396-401.
Trapp, B.D. et al. Axonal transection in the lesions of multiple sclerosis. N Engl J Med 1998; 338: 278-285.
Triaca, V. et al. Presence of nerve growth factor and TrkA expression in the SVZ of EAE rats: evidence for a possible functional significance. Experimental Neurology 2005; 191: 53-64.
Tsunoda, I. et al. Converting relapsing remitting to secondary progressive experimental allergic encephalomyelitis (EAE) by ultraviolet B irradiation. J Neuroimmunol 2005; 160(1-2): 122-34.
Urban, L. et al. Activity of deep dorsal horn neurons in the anaesthetized rat during hyperalgesia of the hindpaw induced by ultraviolet irradiation. Neuroscience 1993; 57(1): 167-72.
Van der Mei, I.A.F. et al. Regional variation in multiple sclerosis prevalence in Australia and its association with ambient ultraviolet radiation. Neuroepidemiology 2001; 20: 168-74.
Valyi-Nagy T, Deshmane S, Dillner A, and Fraser NW. Induction of cellular transcription factors in trigeminal ganglia of mice by corneal scarification, herpes simplex virus type 1 infection, and explantation of trigeminal ganglia. J Virol 1991; 65(8): 4142-52.
van de Meent H, Hamers FP, Lankhorst AJ, Joosten EA, and Gispen WH. Beneficial effects of the melanocortin alpha-melanocyte-stimulating hormone on clinical and neurophysiological recovery after experimental spinal cord injury. Neurosurgery 1997; 40(1): 122-30.
van der Mei IA, Ponsonby AL, Blizzard L, and Dwyer T. Regional variation in multiple sclerosis prevalence in Australia and its association with ambient ultraviolet radiation. Neuroepidemiology 2001; 20(3): 168-74.
van der Mei IA, Ponsonby AL, Dwyer T, Blizzard L, Simmons R, Taylor BV, Butzkueven H, and Kilpatrick T. Past exposure to sun, skin phenotype, and risk of multiple sclerosis: case-control study. BMJ 2003; 327(7410): 316.
Wallengren, J., and Sundler, F. Phototherapy reduces the number of epidermal and CGRP-positive dermal nerve fibres. Acta Derm Venereol 2004; 84: 111-115.
Weinberg, W.C. et al. P21WAF1 control of epithelial cell cycle and cell fate. Crit Rev Oral Biol Med 2002; 13(6): 453-464.
Weiss MD, Luciano CA, Semino-Mora C, Dalakas MC, and Quarles RH. Molecular mimicry in chronic inflammatory demyelinating polyneuropathy and melanoma. Neurology 1998; 51(6): 1738-41.
Wintzen, M. et al. Proopiomelanocortin gene product regulation in keratinocytes. J Invest Dermatol 1996; 106: 673-8.
Wintzen, M. et al. Total body exposure to ultraviolet radiation does not influence plasma levels of immunoreactive beta-endorphin in man. Photodermatol Photoimmunol Photomed 2001; 17: 256-60.
Yarnitsky D, Goor-Aryeh I, Bajwa ZH, Ransil BI, Cutrer FM, Sottile A, and Burstein R. 2003 Wolff Award: Possible parasympathetic contributions to peripheral and central sensitization during migraine. Headache 2003; 43(7): 704-14.
Yavari P, McCulloch PF, and Panneton WM. Trigeminally-mediated alteration of cardiorespiratory rhythms during nasal application of carbon dioxide in the rat. J Auton Nerv Syst 1996; 61(2): 195-200.
Yin P, Luby TM, Chen H, Etemad-Moghadam B, Lee D, Aziz N, Ramstedt U, and Hedley ML. Generation of expression constructs that secrete bioactive alphaMSH and their use in the treatment of experimental autoimmune encephalomyelitis. Gene Ther 2003; 10(4): 348-55.
Zeller, S. et al. Do adolescent indoor tanners exhibit dependency? J Am Acad Dermatol 2006; 54: 589-96.