“NO/ONOO” theory

<a href="https://link.springer.com/article/10.1007/s12035-018-0928-9">A Molecular Neurobiological Approach to Understanding the Aetiology of Chronic Fatigue Syndrome (Myalgic Encephalomyelitis or Systemic Exertion Intolerance Disease) with Treatment Implications | SpringerLink</a>

<h2 class="Heading">Abstract</h2>

Currently, a psychologically based model is widely held to be the basis for the aetiology and treatment of chronic fatigue syndrome (CFS)/myalgic encephalomyelitis (ME)/systemic exertion intolerance disease (SEID). However, an alternative, molecular neurobiological approach is possible and in this paper evidence demonstrating a biological aetiology for CFS/ME/SEID is adduced from a study of the history of the disease and a consideration of the role of the following in this disease: nitric oxide and peroxynitrite, oxidative and nitrosative stress, the blood–brain barrier and intestinal permeability, cytokines and infections, metabolism, structural and chemical brain changes, neurophysiological changes and calcium ion mobilisation. Evidence is also detailed for biologically based potential therapeutic options, including: nutritional supplementation, for example in order to downregulate the nitric oxide-peroxynitrite cycle to prevent its perpetuation; antiviral therapy; and monoclonal antibody treatment. It is concluded that there is strong evidence of a molecular neurobiological aetiology, and so it is suggested that biologically based therapeutic interventions should constitute a focus for future research into CFS/ME/SEID.

</section>

<h2 class="Heading">Keywords</h2>

Chronic fatigue syndrome Myalgic encephalomyelitis Systemic exertion intolerance disease Molecular neurobiology </div>

<h2 class="Heading" tabindex="-1" data-role="collapsible-handle">Introduction</h2>

The disorder variously known inter alia as chronic fatigue syndrome (CFS), myalgic encephalomyelitis (ME) and systemic exertion intolerance disease (SEID) has a phenotype of unknown aetiology, whilst there is considerable controversy over the most appropriate treatment approach(es). In this in-depth review, we bring together the results of research into the molecular neurobiological mechanisms which underpin CFS/ME/SEID, thereby providing helping to inform an evidence-based approach to its treatment.

After briefly discussing the history and definition of this disorder, we consider a wide variety of molecular neurobiological factors and we then describe an evidence-based approach to the treatment of CFS/ME/SEID.

Nitric Oxide and Peroxynitrite

The NO/ONOO− Cycle

Pall has described a vicious cycle which is initiated by nitric oxide (NO; official name—nitrogen monoxide) and peroxynitrite (ONOO−; official name—oxoperoxonitrate (1−)) [<a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR15">15</a>, <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR16">16</a>]. Although Pall refers to this as the NO/ONOO− cycle, it could also be referred to as the •NO/ONOO− cycle since the nitrogen in NO has an unpaired electron in the π*(2p) antibonding orbital, meaning that the nitric oxide is a free radical. While the whole cycle needs to be considered in any individual, interlocking cyclic components of it can be scrutinised. Some important aspects of such components of the •NO/ONOO− cycle are now described and are illustrated in Fig. <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#Fig1">1</a>.

<a href="https://media.springernature.com/original/springer-static/image/art%3A10.1007%2Fs12035-018-0928-9/MediaObjects/12035_2018_928_Fig1_HTML.gif">Open image in new window</a>

Fig. 1

Key aspects of the nitric oxide–peroxynitrite cycle. See text for details (partly based upon figure 1.2 in reference [<a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR15">15</a>])

In terms of energy metabolism, it is noted first that excessive nitric oxide produces more superoxide and then a depletion of adenosine triphosphate (ATP). This leads to activation of N-methyl-d-aspartate (NMDA) through its receptors. This can result in heightened intracellular calcium, which in turn will trigger increased activity of endothelial nitric oxide synthase (eNOS) and of neuronal nitric oxide synthase (nNOS) and perpetuation of increased nitric oxide. Furthermore, superoxide can activate transient receptor potential (TRP) channels and NMDA. These work principally through vanilloid TRPV1 receptors, which are activated in multiple chemical sensitivity (MCS) [<a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR17">17</a>, <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR18">18</a>, <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR19">19</a>, <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR20">20</a>]. When TRP receptors, active NMDA receptors and calcium are increased, this results in voltage-gated calcium channels being activated, and Pall has implicated this mechanism in food and chemical sensitivity [<a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR17">17</a>, <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR18">18</a>, <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR19">19</a>, <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR20">20</a>].

Another cyclic component of the overall •NO/ONOO− cycle involves the reaction with nitric oxide with superoxide to form peroxynitrite, as follows:

∙NO+O2∙−→ONOO−∙NO+O2∙−→ONOO−

Thus, elevated nitric oxide leads to increased superoxide and peroxynitrite. This leads to oxidative stress and upregulation of the protein complex nuclear factor kappa-light-chain enhancer of activated B cells (NFκB), with many pro-inflammatory cytokines (PICs) being induced. These in turn increase the activity of inducible nitric oxide synthase (iNOS), which increases nitric oxide. Oxidative stress itself can also lead to increased calcium, which triggers nitric oxide production, while NFκB and PICs can be provoked by infections [<a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR21">21</a>]. In addition to superoxide provoking a depletion of ATP (see above), peroxynitrite can also do so [<a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR15">15</a>, <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR16">16</a>].

Peroxynitrite oxidises the electron carrier tetrahydrobiopterin (BH4), which is derived from the cofactor biopterin. BH4 acts as a NOS cofactor as well as playing a role in the degradation of aromatic amino acids and the biosynthesis of catecholamine and serotonin/melatonin [<a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR22">22</a>]. For example, in the hydroxylation of phenylalanine to tyrosine (catalysed by phenylalanine hydroxylase), BH4 is converted into quinonoid dihydrobiopterin; the latter is reduced, by nicotinamide adenine dinucleotide (NADH), back into BH4 (catalysed by dihydropteridine reductase).

</div>

</section>

</div>

Pall has adduced evidence that various components of his NO/ONOO− cycle are abnormal in CFS/ME/SEID [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR21">21</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR22">22</a>]. First, the following stressors, which have been implicated in the initiation of CFS/ME/SEID, can lead to increased nitric oxide levels: infection by viruses, bacteria and protozoa; carbon monoxide exposure; ciguatoxin poisoning; physical trauma; psychological stress; and exposure to ionising radiation [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR22">22</a>]. Second, many studies have reported that CFS/ME/SEID is associated with elevated levels of markers of oxidative stress, nitric oxide levels and PICs (reviewed by reference [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR22">22</a>]).

Chronic activation of NFκB is hypothesised to occur in CFS/ME/SEID (see Fig. <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#Fig1">1</a>); in 2011, Hoeck and Pall suggested that vitamin D3 (1,25(OH)2D3) supplementation may be helpful in this disease owing to the ability of active vitamin D3 to repress activation of NFκB, via binding to vitamin D receptor [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR23">23</a>]. To date, there are no published trial data testing this suggestion. However, 3 years later, Witham and colleagues independently reported that in CFS/ME patients, the level of vitamin D3 is inversely correlated with markers of inflammation, oxidative stress and cardiovascular risk [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR24">24</a>]. NFκB activation is likely to occur in response to ionising radiation, via free radical and pro-oxidant molecules, and Pall has hypothesised that this may constitute part of the mechanism for symptoms of post-radiation syndrome, which is a CFS/ME/SEID-like disorder [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR25">25</a>].

In 2003, Smirnova and Pall reported that the serum protein carbonyl levels are elevated in CFS patients compared with controls, while there is no significant difference in total protein levels [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR26">26</a>]. Since protein carbonyl levels index protein oxidation, this finding is consistent with increased oxidative stress in CFS/ME and therefore offers support for Pall’s NO/ONOO− cycle model (see Fig. <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#Fig1">1</a>).

Both CFS/ME/SEID and cardiac failure are associated with fatigue [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR27">27</a>]. In 2013, Pall published extensive evidence, including 34 mechanisms, pointing to the NO/ONOO− cycle as being central to the cause of cardiac failure [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR28">28</a>]. There is also evidence of an association between CFS/ME/SEID and multiple chemical sensitivity (MCS) [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR29">29</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR30">30</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR31">31</a>]. Pall has suggested that MCS may be a response to the index chemicals causing increased NMDA activity (see Fig. <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#Fig1">1</a>) [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR17">17</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR19">19</a>]. Again, in 2013, Pall showed how volatile chemicals may act as toxicants via both transient receptor potential ankyrin 1 (TRPA1) and TRPV1 receptors; the latter are part of the NO/ONOO− cycle (see Fig. <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#Fig1">1</a>) [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR18">18</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR32">32</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR33">33</a>].

</section>

</div>

</section>

<h2 class="Heading" tabindex="-1" data-role="collapsible-handle">Oxidative and Nitrosative Stress</h2>

Fukuda and colleagues recently showed that, at rest, measures of oxidative stress were higher, and measures of biological antioxidant potential lower, in CFS patients than in healthy volunteers [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR34">34</a>]. These may well adversely impact on lipoprotein-based cellular signalling; Morris and colleagues have pointed out that oxidative and nitrosative stress (O&NS) affects the following lipid-based signalling mechanisms: the post-translational S-palmitoylation modification; the functions of omega-3 polyunsaturated fatty acids (PUFAs); and the functions of membrane/lipid rafts [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR35">35</a>].

A recent study of the use of nutraceuticals with actions against inflammatory, O&NS in 76 CFS/ME patients was carried out by Maes and Leunis [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR36">36</a>]. At baseline, the patients had abnormal autoimmune responses. Improved clinical outcome was associated with a reduction in autoimmune responses to oxidative specific epitopes, such as malondialdehye (MDA) and phosphatidyl inositol, rather than to a reduction in nitroso-adducts such as nitroso-arginine, nitroso-cysteinyl and nitroso-tryptophan [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR36">36</a>].

</div>

</section>

<h2 class="Heading" tabindex="-1" data-role="collapsible-handle">Blood–Brain Barrier Damage and Intestinal Permeability</h2>

When the blood–brain barrier is damaged, circulating antibodies that cross-react with neurological tissues can infiltrate the brain and nervous tissue, with the potential destruction of neurological tissues. The cycle of neuroautoimmunity can begin with a breach of the gastrointestinal and/or blood–brain barriers. It is also important to bear in mind that there exist gaps in the blood–brain barrier to allow hypothalamic-hormonal control via biofeedback; therefore, the (neuro-)lymphatic/glymphatic drainage is important for brain health and possibly for CFS/ME/SEID [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR37">37</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR38">38</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR39">39</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR40">40</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR41">41</a>].

<div id="Par23" class="Para">Cross and colleagues have carried out murine experiments on acute experimental autoimmune encephalomyelitis (which serves as an animal model for human multiple sclerosis) which show that nitrotyrosine is found in the CNS after exposure to peroxynitrite [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR42">42</a>]. The formation of nitrotyrosine under physiological conditions is indicative of peroxynitrite damage of proteins (including the amino acid tyrosine) [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR42">42</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR43">43</a>]. The molecular mechanism is likely to involve the formation of the intermediate nitrosoperoxycarbonate (O=NOOCO2−) by the following reaction:

<div class="MathJax_Display"><span id="MathJax-Element-2-Frame" class="MathJax" style="box-sizing: border-box; display: block; font-style: normal; font-weight: normal; line-height: normal; font-size: 17px; text-indent: 0px; text-align: left; text-transform: none; letter-spacing: normal; word-spacing: normal; word-wrap: normal; white-space: nowrap; float: none; direction: ltr; max-width: 100%; max-height: none; min-width: 0px; min-height: 0px; border: 0px; padding: 0px; margin: 0px; width: 727px; overflow: auto hidden; position: relative;" tabindex="0" data-mathml="<math xmlns="http://www.w3.org/1998/Math/MathML" display="block"><msup><mrow class="MJX-TeXAtom-ORD"><mrow class="MJX-TeXAtom-ORD"><mi mathvariant="normal">O</mi><mi mathvariant="normal">N</mi><mi mathvariant="normal">O</mi><mi mathvariant="normal">O</mi></mrow></mrow><mrow class="MJX-TeXAtom-ORD"><mo>&#x2212;</mo></mrow></msup><mo>+</mo><msub><mrow class="MJX-TeXAtom-ORD"><mrow class="MJX-TeXAtom-ORD"><mi mathvariant="normal">C</mi><mi mathvariant="normal">O</mi></mrow></mrow><mn>2</mn></msub><mo stretchy="false">&#x2192;</mo><msup><mrow class="MJX-TeXAtom-ORD"><msub><mrow class="MJX-TeXAtom-ORD"><mrow class="MJX-TeXAtom-ORD"><mi mathvariant="normal">O</mi><mi mathvariant="normal">N</mi><mi mathvariant="normal">O</mi><mi mathvariant="normal">O</mi></mrow><mrow class="MJX-TeXAtom-ORD"><mi mathvariant="normal">C</mi><mi mathvariant="normal">O</mi></mrow></mrow><mn>2</mn></msub></mrow><mrow class="MJX-TeXAtom-ORD"><mo>&#x2212;</mo></mrow></msup></math>">ONOO−+CO2→ONOOCO2−ONOO−+CO2→ONOOCO2−followedbyhomolytic</div>

</div>

fission of the nitrosoperoxycarbonate to two reactive radicals:

<div class="MathJax_Display"><span id="MathJax-Element-3-Frame" class="MathJax" style="box-sizing: border-box; display: block; font-style: normal; font-weight: normal; line-height: normal; font-size: 17px; text-indent: 0px; text-align: left; text-transform: none; letter-spacing: normal; word-spacing: normal; word-wrap: normal; white-space: nowrap; float: none; direction: ltr; max-width: 100%; max-height: none; min-width: 0px; min-height: 0px; border: 0px; padding: 0px; margin: 0px; width: 727px; overflow: auto hidden; position: relative;" tabindex="0" data-mathml="<math xmlns="http://www.w3.org/1998/Math/MathML" display="block"><msup><mrow class="MJX-TeXAtom-ORD"><msub><mrow class="MJX-TeXAtom-ORD"><mrow class="MJX-TeXAtom-ORD"><mi mathvariant="normal">O</mi><mi mathvariant="normal">N</mi><mi mathvariant="normal">O</mi><mi mathvariant="normal">O</mi><mi mathvariant="normal">C</mi><mi mathvariant="normal">O</mi></mrow></mrow><mn>2</mn></msub></mrow><mrow class="MJX-TeXAtom-ORD"><mo>&#x2212;</mo></mrow></msup><mo stretchy="false">&#x2192;</mo><msup><mrow class="MJX-TeXAtom-ORD"></mrow><mrow class="MJX-TeXAtom-ORD"><mo>&#x2219;</mo></mrow></msup><msub><mrow class="MJX-TeXAtom-ORD"><mrow class="MJX-TeXAtom-ORD"><mi mathvariant="normal">N</mi><mi mathvariant="normal">O</mi></mrow></mrow><mn>2</mn></msub><mo>+</mo><msup><mrow class="MJX-TeXAtom-ORD"><msub><mrow class="MJX-TeXAtom-ORD"><mrow class="MJX-TeXAtom-ORD"><mi mathvariant="normal">C</mi><mi mathvariant="normal">O</mi></mrow></mrow><mn>3</mn></msub></mrow><mrow class="MJX-TeXAtom-ORD"><mo>&#x2219;</mo><mo>&#x2212;</mo></mrow></msup></math>">ONOOCO2−→∙NO2+CO3∙−ONOOCO2−→∙NO2+CO3∙−</div>

</div>

Following oxidation of tyrosine by the second of these two radicals (the carbonate radical, CO3•−), the product, tyr-O•, then combines with the first radical, •NO2, to form nitrotyrosine [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR43">43</a>]. Since nitrotyrosine is found within the CNS following peripheral exposure to peroxynitrite, it seems reasonable to conclude that peroxynitrite damages the blood–brain barrier; given the evidence favouring increased peroxynitrite activity in CFS/ME/SEID, blood–brain barrier damage in this disorder may therefore be expected [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR44">44</a>].

As alluded to earlier, viral infections and PICs are associated with increased •NO levels, often as a result of increased iNOS expression [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR45">45</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR46">46</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR47">47</a>], which in turn leads to increased peroxynitrite levels. Again, by the above mechanism, this may lead to blood–brain barrier damage [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR42">42</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR44">44</a>]. Evidence of the association of infections and PICs, on the one hand, with CFS/ME/SEID, on the other hand, is given in later sections of this paper.

Staines and colleagues have hypothesised that immunopathology of the cerebrospinal perivascular compartment may occur in CFS/ME/SEID, with the particular involvement of the vasoactive neuropeptides pituitary adenylate cyclase-activating polypeptide (PACAP) and vasoactive intestinal peptide (VIP) [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR48">48</a>]. Breaching of these barriers may result in the damaging effects of Th1 and Th17 lymphocytes, as well as antibodies that can target and damage neurones and tissues [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR49">49</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR50">50</a>]. Furthermore, elevated cerebrospinal fluid (CSF) levels of the PIC tumour necrosis factor-alpha (TNF-α) have been reported in CFS patients compared with non-CFS controls [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR51">51</a>].

Environmental exposures of sufficient magnitude, including gut dysbiosis, may also trigger opening of tight junctions in the gastrointestinal tract, leading to the entry into the circulation of lipopolysaccharide (LPS) molecules, as well as undigested molecules and bacterial toxins; similarity between some gastrointestinal endothelial tissue proteins and certain proteins of the blood–brain barrier may lead to ‘leaky brain’ [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR52">52</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR53">53</a>]. In a rigorously controlled study of 50 CFS/ME patients compared with 50 matched healthy controls, CFS/ME was indeed found to be associated with gut dysbiosis [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR54">54</a>]. The importance of this phenomenon in CFS/ME/SEID needs further study.

</div>

</section>

<h2 class="Heading" tabindex="-1" data-role="collapsible-handle">Cytokines and Infections</h2>

The finding of increased CSF TNF-α in CFS has been mentioned above [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR51">51</a>]. Several other studies have confirmed that CFS/ME/SEID is associated with increased levels of cytokines, particularly those which are pro-inflammatory. Four further recent examples are described here.

Using a high-throughput 51-multiplex array, Montoya and colleagues studied serum cytokines in 192 ME/CFS patients and 392 healthy controls in the important Stanford Myalgic Encephalomyelitis Initiative [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR55">55</a>]. The following cytokines showed a significant elevation which correlated with disease severity: CCL11 (Eotaxin-1), CXCL1 (GROα), CXCL10 (IP-10), IFN-γ, IL-4, IL-5, IL-7, IL-12p70, IL-13, IL-17F, leptin, G-CSF, GM-CSF, LIF, NGF, SCF and TGF-α; of these, 13 are PICs and are likely to contribute to symptoms [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR55">55</a>].

Using a 51-multiplex array, Hornig and colleagues studied CSF cytokines in 32 patients and 19 matched controls; their results pointed to a disturbance in IL-1 signalling [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR56">56</a>]. In an earlier study, Horning and fellow workers also found evidence of cytokine activation and dissociation of inter-cytokine regulatory networks early in the course of CFS/ME, with illness duration being better correlated with cytokine changes than illness severity [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR57">57</a>].

A study by Milrad and colleagues showed that poor sleep quality in CFS/ME patients is associated with higher levels of the PICs TNF-α, IL-1β and IL-6 [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR58">58</a>].

These studies suggest that CFS/ME/SEID has a dynamic immunopathology with CNS immune activation and a shift to the Th2 pattern seen in autoimmunity [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR56">56</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR57">57</a>]. Furthermore, neuroinflammation, together with O&NS and neural sensitivity, can be formulated into a robust model of the aetiopathology of CFS/ME/SEID involving sustained glial activation [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR59">59</a>].

As mentioned earlier, increased PIC levels are associated with infections. It is therefore noteworthy that Underhill has suggested that CFS/ME is an infectious disease, on the basis of clinical, epidemiological and immunological evidence [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR60">60</a>]. The following evidence particularly supports the role of Epstein–Barr virus (EBV) or human herpesvirus-4 (HHV-4), which can cause infectious mononucleosis, in this regard.

At 6 months, 1 year and 2 years after infection, of 301 adolescents, aged between 12 and 18 years, with infectious mononucleosis, 13, 7 and 4%, respectively, met CFS criteria in a study by Katz and colleagues [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR61">61</a>]. While CFS/ME/SEID patients and controls have been found to have similar IgG antibody response patterns to EBV, patients show increased IgG reactivity to an EBNA-6 repeat sequence and also to EBNA-6 protein; it may be that homologous sequences of human proteins containing this sequence could give rise to antigenic mimicry [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR62">62</a>].

It is also interesting to note that there is evidence that CFS/ME/SEID patients suffer from sensitised fatigue pathways [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR63">63</a>], as this may be related to the spread of reactivated EBV to ectopic lymphoid aggregates [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR64">64</a>].

Besides EBV, Montoya and colleagues have suggested that HHV-6 may also be an infectious trigger for CFS/ME [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR65">65</a>]. Importantly, Pantry and colleagues have shown that chromosomal integrated HHV-6 can be associated with persistent infection with exogenous HHV-6, which in turn may be associated with neurological symptomology [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR66">66</a>].

</div>

</section>

<h2 class="Heading" tabindex="-1" data-role="collapsible-handle">Metabolism</h2>

There is now clear evidence of metabolic dysfunction in CFS/ME/SEID. As mentioned above, CFS/ME/SEID is associated with the presence of chronic O&NS, low-grade inflammation and impairment of the production of heat shock proteins [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR67">67</a>]. Naviaux and colleagues studied 45 patients and 39 age- and sex-matched controls using targeted, broad-spectrum metabolomics and found evidence that, in this disorder, 20 examined metabolic pathways were abnormal, including those involved in mitochondrial and peroxisomal metabolism and those involved in the metabolism of branched-chain amino acids, cholesterol, phospholipids, purines, pyrroline-5-carboxylate, riboflavin, sphingolipid, as well as microbiome metabolism [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR68">68</a>]. In a more recent metabolic profiling study, abnormalities in purine, pyrimidine and amino acid metabolic pathways were found (including adenosine diphosphate and ATP) as well as in fatty acid and lipid metabolism [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR69">69</a>]. There is also evidence of impaired pyruvate dehydrogenase function [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR70">70</a>], which is consistent with a shift from oxidative phosphorylation to the production of lactic acid in this disease.

<div id="Par38" class="Para">In their comprehensive metabolomic analyses of plasma samples from 46 CFS patients and 47 age- and sex-matched controls, Yamano and fellow co-workers reported group differences in both the Kreb’s (tricarboxylic or citric acid) cycle, with lower citrate, lower cis-aconitate, lower isocitrate and lower malate in the patient group (illustrated in Fig. <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#Fig2">2</a>), and in the urea cycle, with the patient group showing higher ornithine, lower citrulline and lower urea levels (see Fig. <a href="https://link.springer.com/article/10.1007/s12035-018-0928-9#Fig3">3</a>) [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR71">71</a>].

<div id="MO2" class="MediaObject"><a href="https://media.springernature.com/original/springer-static/image/art%3A10.1007%2Fs12035-018-0928-9/MediaObjects/12035_2018_928_Fig2_HTML.gif" target="_blank" rel="noopener">Open image in new window</a><img src="/f6410e3e-95e3-4d4f-ae74-16c4f7eb977c.gif" /></div>

<div class="CaptionContent">Fig. 2

Metabolic changes in Kreb’s cycle reported in CFS/ME/SEID (based on data in reference [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR71">71</a>])

</div>

</figcaption>

</figure>

<div id="MO3" class="MediaObject"><a href="https://media.springernature.com/original/springer-static/image/art%3A10.1007%2Fs12035-018-0928-9/MediaObjects/12035_2018_928_Fig3_HTML.gif" target="_blank" rel="noopener">Open image in new window</a><img src="/75c17c25-9a1b-4217-8e94-1c8497d97d5f.gif" /></div>

<div class="CaptionContent">Fig. 3

Metabolic changes in the urea cycle reported in CFS/ME/SEID (based on data in reference [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR71">71</a>])

</div>

</figcaption>

</figure>

</div>

</section>

<h2 class="Heading" tabindex="-1" data-role="collapsible-handle">Brain Changes</h2>

One of the largest voxel-based morphometry brain MRI cross-sectional studies at a 3-T magnetic field strength was that carried out by Puri and colleagues, which compared 26 CFS/ME/SEID patients with 26 age- and sex-matched controls [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR72">72</a>]. The patient group had reduced grey matter in the following three regions. First, the occipital lobes, including both poles, the superior division of the left lateral cortex and the supracalcrine cortex also on the left side of the brain. Second, the angular gyrus of the right hemisphere. Third, the posterior division of the left parahippocampal gyrus. Reduced white matter was found in the patient group in the left occipital lobe [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR72">72</a>]. These particular findings are consistent with the symptom of impaired memory which is well described in CFS/ME/SEID, and they also point to the possibility of visual processing abnormalities and the possibility of discrepancies existing between intentions and corresponding actions [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR72">72</a>]. This has been described as just one of several structural neuroimaging studies which have shown that CFS/ME/SEID is associated with significant neuroanatomical changes.

Furthermore, as Shan and colleagues found in their longitudinal MRI study of 15 patients and 10 controls, who were scanned (at 1.5 T) at baseline and 6-year follow-up, significant neuroanatomical changes occur over time [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR73">73</a>]. In the case of this study, there was a decrease in white matter in the left inferior fronto-occipital fasciculus, which was not seen in the control group [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR73">73</a>]. It should be noted, however, that not all recent longitudinal MRI studies have shown evidence of changes. Notably, the longitudinal study of Perrin and colleagues proved negative [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR74">74</a>].

The chemistry of the brain has also been studied in this disorder, using magnetic resonance spectroscopy (MRS). The first systematic proton-MRS study of CFS/ME patients compared with age- and sex-matched controls was that of Puri and colleagues, who in 2002 reported that CFS was associated with a relative increase of choline-containing compounds in the occipital cortex [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR75">75</a>]. The following year, Chaudhuri and colleagues similarly systematically studied the left basal ganglia of CFS/ME patients and age- and sex-matched controls using proton-MRS; they also found a relative increase of choline-containing compounds in this part of the brain [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR76">76</a>]. These MRS findings may point to a problem with phospholipid metabolism, increased membrane turnover owing to gliosis or changes in intra-membrane signalling [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR75">75</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR76">76</a>]. They may also be consistent with Perrin’s much earlier prediction of increased central levels of acetylcholine [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR77">77</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR78">78</a>].

Thus, there is good evidence of the existence of neuroanatomical and neurochemical changes in CFS/ME/SEID. This is also ample evidence of neurophysiological changes, which are discussed in the next section.

</div>

</section>

<h2 class="Heading" tabindex="-1" data-role="collapsible-handle">Neurophysiological Changes</h2>

Rasouli and colleagues have recently published evidence that, compared with healthy controls, CFS/ME patients show poorer gross motor function, assessed by taking longer on a reaction time task, and also poorer fine motor function, assessed using a pegboard task [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR79">79</a>].

Increased symptom intensity in CFS has been reported to occur as a result of neuromuscular strain, with such strain (as opposed to sham strain) being associated with increased somatic pain and problems with concentration [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR80">80</a>].

A whole-genome sequencing study in 95 CFS/ME patients and 77 age- and sex-matched controls, by Johnston and colleagues, has shown preliminary evidence of a higher prevalence of a SNP in the adrenergic receptor α1 (ADRA1A) in the patient group [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR81">81</a>]. Further study of adrenergic receptors in CFS/ME/SEID is clearly indicated.

</div>

</section>

<h2 class="Heading" tabindex="-1" data-role="collapsible-handle">Calcium Ion Mobilisation</h2>

Recent evidence from the work of Nguyen and colleagues points to the importance of natural killer (NK) cell calcium ion mobilisation in CFS/ME/SEID [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR82">82</a>]. CD56bright CD16dim/− NK cells are important in immunosurveillance and produce relatively large amounts of cytokines. In contrast, CD56dim CD16+ NK cells are cytotoxic, killing infected cells, tumour cells and cells which are ‘missing self’. Compared with healthy controls, immunosurveillance NK cells (of the first type just mentioned) from CFS/ME patients have been found to have reduced transient receptor potential melastatin subfamily 3 ion channels, which are involved in calcium ion signalling [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR82">82</a>]. When CFS/ME-patient NK cells of this type were stimulated with pregnenolone sulphate (sulfate), they showed increased calcium ion flux compared with the same NK cell type from the control group [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR82">82</a>].

While it is too early to speculate on the implications of the above results, they point to the intriguing possibility that calcium ion mobilisation in NK cells may well be important in the pathogenesis of CFS/ME/SEID. This is consistent with the importance of intracellular calcium ion concentration in Pall’s NO/ONOO− cycle [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR15">15</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR18">18</a>], mentioned earlier in this paper. Moreover, Pall has also pointed out the association of CFS/ME/SEID with MCS, with both disorders being associated with increased NMDA activity [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR18">18</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR19">19</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR20">20</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR22">22</a>]. Furthermore, our group have recently found that desensitisation treatment for chemical and food sensitivities using low-dose immunotherapy ascertained by provocation neutralisation is itself associated with reduced influx of calcium ions into lymphocytes [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR83">83</a>]. Thus, it seems reasonable to suggest that this particular therapeutic intervention should be formally studied in a trial in CFS/ME/SEID patients. We now turn to treatment approaches which have already been the subject of published studies.

</div>

</section>

<h2 class="Heading" tabindex="-1" data-role="collapsible-handle">Treatment</h2>

In this section, published evidence of the efficacy of some biological, non-psychological, non-psychiatric interventions for CFS/ME/SEID are described. (One physician, himself bedridden with ME, has even argued that graded exercise therapy and cognitive behavioural therapy may be harmful to patients [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR84">84</a>].)

From the description of the NO/ONOO− cycle earlier in this paper, it follows that downregulation of this cycle might be expected to be of therapeutic value in CFS/ME/SEID. One candidate to achieve such downregulation is vitamin B12, which is available over the counter in different forms, including as hydroxocobalamin, cyanocobalamin and methylcobalamin. In 1995, using porcine endothelial cells from the aorta, Rochelle and colleagues published evidence favouring the existence of a redox reaction between reduced cobalamin and NO, with binding of NO occurring in a reversible manner to oxidised cobalamin [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR85">85</a>]. Since then, several studies, both in vitro and in vivo, have established the NO scavenger role of cobalamin [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR15">15</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR86">86</a>]. This may help explain the results of an early double-blind cross-over trial in patients complaining of tiredness (the study was from the 1970s, well before the publication of more recent operational criteria for the diagnosis of CFS/ME/SEID), in which twice-weekly injections of hydroxocobalamin, for a fortnight, were associated with improved well-being which persisted for at least a month [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR87">87</a>]. At the time of writing, there have been no publications of any similar trials in operationally defined CFS/ME/SEID, although anecdotal reports suggest that there may be a benefit of this vitamin in such patients, with this benefit not necessarily being associated with a pre-existing vitamin B12 deficiency [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR15">15</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR88">88</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR89">89</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR90">90</a>].

The lipophilic coenzyme known as coenzyme Q10 (CoQ10) has important roles in ATP generation, inflammatory cascade inhibition and apoptosis prevention [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR91">91</a>]. CoQ10 can act as an antioxidant which can inhibit the oxidation of DNA, lipids and proteins [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR92">92</a>]. It can also prevent the initiation and propagation of lipid peroxidation, including by means of the recycling action of NAD(P)H:(quinone acceptor) oxidoreductase 1 activity [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR92">92</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR93">93</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR94">94</a>]. These actions occur in vivo in mammals; murine experiments have shown that dietary supplementation is associated with reduced lipid peroxidation [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR95">95</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR96">96</a>]. This occurs also in membranes of mitochondria located both peripherally and in the CNS and is associated with improved mitochondrial function, again both peripherally and centrally [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR97">97</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR98">98</a>]. Therefore, it would seem reasonable to propose that supplementation with CoQ10 and NADH could be beneficial in CFS/ME/SEID. In a recent double-blind 8-week study by Castro-Marrero and colleagues of 80 CFS patients who were randomised to receive either this supplement combination or a matching placebo, the active group showed a reduced maximum cardiac rate during a cycle ergometer test at the end of the study (compared with baseline) and indeed also a reduced perception of fatigue [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR99">99</a>]. It is of interest to note that another mitoprotective dietary intervention which has been proposed for CFS/ME/SEID patients is caloric restriction, but this has not yet been the subject of a published trial in this disease [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR100">100</a>].

The mainly pineal neurohormone melatonin (N-acetyl-5-methoxytryptamine) has important antioxidant, neuroprotective and immunomodulatory properties and may help prevent or treat bacterial and viral infections [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR101">101</a>]. Furthermore, the normal circadian rhythms of fatigue and of urinary melatonin levels covary [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR102">102</a>], while administration of this indole to healthy volunteers has been found to be associated with a reduction in self-rated fatigue or tiredness [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR103">103</a>]. It seems reasonable, therefore, to suggest that melatonin supplementation may be of therapeutic value in CFS/ME/SEID patients. While some earlier pilot studies in CFS/ME/SEID gave essentially negative results, or, in one case, there was evidence of high nocturnal melatonin levels therefore suggesting that it would be inappropriate to administer melatonin, a few more recent studies have given positive findings [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR104">104</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR105">105</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR106">106</a>, <a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR107">107</a>]. On balance, the evidence suggests that it would be appropriate to conduct larger, randomised, double-blind trials of melatonin supplementation in CFS/ME/SEID patients [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR108">108</a>]. Meanwhile, it would appear sensible to advise CFS/ME/SEID patients, and indeed most people whether ill or not, to avoid the prolonged evening use of electronic gadgets that employ light-emitting diode screens giving off relatively high levels of blue light (with a wavelength of approximately 470 nm), as exposure of healthy volunteers to such light for just half an hour, starting at 8 p.m., is associated with a strong suppression of nocturnal melatonin production (of over 90%) [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR109">109</a>].

The association of CFS/ME/SEID with EBV or HHV-4 infection has been mentioned above. The use of antiviral treatment for infectious mononucleosis is controversial and is not currently routinely recommended [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR110">110</a>]. However, in the mid-1990s a small, double-blind, placebo-controlled, phase III cross-over study of antiviral treatment in CFS patients showed promising results, with continued improvement in symptomology with up to 18 months’ treatment [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR111">111</a>]. It could therefore be argued that a larger trial in CFS/ME/SEID would be in order. Meanwhile, it may be prudent to consider the role of the intestinal microbiota and virome, with the use of suitable supplementation (although again evidence from clinical trials is currently lacking) [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR112">112</a>].

Twenty nine CFS/ME patients took part in an open-label study of the monoclonal anti-CD20 antibody rituximab, administered as a couple of infusions a fortnight apart followed by maintenance treatments [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR113">113</a>]. Use of this monoclonal antibody is associated with depletion of B lymphocytes; in this study, clinical improvements were observed in almost two-thirds of the participants, with remission being maintained at the time of 3-year follow-up in a majority of the responders [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR113">113</a>]. At the time of writing, it is not yet known how to differentiate responders from non-responders, although it has been noted that responders have lower levels of baseline serum IgG [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR114">114</a>]. Furthermore, in late 2017, it was reported that preliminary results from a phase III trial of rituximab in CFS/ME may have been associated with negative findings; the authors feel that it would be prudent to await formal publication of the full results before commenting further.

Unsurprisingly, in light of the above findings, it has recently been suggested that rehabilitation for CFS/ME/SEID patients should be extended from a narrow psychologically based domain and become multidisciplinary, including for example exercise physiologists and physiotherapists [<a title="View reference" href="https://link.springer.com/article/10.1007/s12035-018-0928-9#CR115">115</a>].

</div>

</section>

<h2 class="Heading" tabindex="-1" data-role="collapsible-handle">Conclusions and Future Directions</h2>

Strong evidence has been presented which points to a molecular neurobiological aetiology of CFS/ME/SEID. Accordingly, it is suggested that biologically based therapeutic interventions should constitute a focus for future research. As has been seen above, preliminary trial data already point to the efficacy of such an approach.

</div>

</section>

“NO/ONOO” theory

Attachments