PON1 Status in Relation to Gulf War Illness: Evidence of Gene–Exposure Interactions from a Multisite Case–Control Study…, 2024, Steele, Klimas+

[SNT Gatchaman]

PON1 Status in Relation to Gulf War Illness: Evidence of Gene–Exposure Interactions from a Multisite Case–Control Study of 1990–1991 Gulf War Veterans
Steele, Lea; Furlong, Clement E.; Richter, Rebecca J.; Marsillach, Judit; Janulewicz, Patricia A.; Krengel, Maxine H.; Klimas, Nancy G.; Sullivan, Kimberly; Chao, Linda L.

BACKGROUND
Deployment-related neurotoxicant exposures are implicated in the etiology of Gulf War illness (GWI), the multisymptom condition associated with military service in the 1990–1991 Gulf War (GW). A Q/R polymorphism at position 192 of the paraoxonase (PON)-1 enzyme produce PON1192 variants with different capacities for neutralizing specific chemicals, including certain acetylcholinesterase inhibitors.

METHODS
We evaluated PON1192 status and GW exposures in 295 GWI cases and 103 GW veteran controls. Multivariable logistic regression determined independent associations of GWI with GW exposures overall and in PON1192 subgroups. Exact logistic regression explored effects of exposure combinations in PON1192 subgroups.

RESULTS
Hearing chemical alarms (proxy for possible nerve agent exposure) was associated with GWI only among RR status veterans (OR = 8.60, p = 0.014). Deployment-related skin pesticide use was associated with GWI only among QQ (OR = 3.30, p = 0.010) and QR (OR = 4.22, p < 0.001) status veterans. Exploratory assessments indicated that chemical alarms were associated with GWI in the subgroup of RR status veterans who took pyridostigmine bromide (PB) (exact OR = 19.02, p = 0.009) but not RR veterans who did not take PB (exact OR = 0.97, p = 1.00). Similarly, skin pesticide use was associated with GWI among QQ status veterans who took PB (exact OR = 6.34, p = 0.001) but not QQ veterans who did not take PB (exact OR = 0.59, p = 0.782).

CONCLUSIONS
Study results suggest a complex pattern of PON1192 exposures and exposure–exposure interactions in the development of GWI.

Link | PDF (International Journal of Environmental Research and Public Health) [Open Access]

 

[Hutan]

Introduction

Gulf War illness (GWI) is a chronic, multisymptom, multi-system disorder estimated to affect approximately 30% of the nearly 700,000 veterans who served in the 1990–1991 Gulf War (GW)

population and clinical studies have consistently identified only a limited number of deployment-related exposures as the most significant risk factors for GWI [1,9]. These include the prolonged use of pesticides and insect repellants [4,8], exposure to nerve agents [10–13], and the use of pyridostigmine bromide (PB), a carbamate drug used as prophylaxis to protect troops from potential Iraqi nerve agent attacks [14,15]. Many of these prominent GWI risk factors exert toxic effects by acting acutely as acetylcholinesterase inhibitors (AChEis), reducing the breakdown of the neurotransmitter acetylcholine in the nervous system [16,17].

Paraoxonase (PON)-1 is a human enzyme capable of hydrolyzing the active metabo- lites (oxons) of a number of organophosphorus (OP) and other xenobiotic substrates [20]. Polymorphisms in the PON1192 coding region (Q192R) can influence the catalytic efficiency of PON1 and affect its ability to hydrolyze different OP compounds [21–23]. For example, in vitro studies suggest that the PON1192 Q alloform is more effective than the PON1192 R alloform in hydrolyzing sarin (a nerve agent) [24], while in vivo knock-out mice studies clearly indicate that the PON1192 R alloform is more effective than the PON1192 Q alloform in protecting against chlorpyrifos oxon exposures (toxic oxon forms of OP compounds) [25].

A limited number of studies have evaluated PON1192 genotypes and alloform in relation to GWI [26–28]. Haley and colleagues [26,27] reported that veterans with the PON1192 R allele were at increased risk of having GWI/Haley syndromes, particularly if they were in locations associated with nerve agent exposures or heard chemical alarms during the war [26,27]. However, these studies did not consider the effects of other GW exposures and PON1 variants.

The current study sought to evaluate GWI risk among GW veterans in relation to PON1 status, defined by functional PON1192 genotype and activity level [29], in a multisite GWI case–control sample. We focused on PON1 status because what determines whether PON1 will protect against a given OP exposure is the catalytic efficiency of PON1 in detoxicating that specific OP compound and the PON1 activity level, which can vary at least 15-fold within a given genotype (Q/Q, Q/R, or R/R) [29]. We also assessed the associations of GWI with a range of GW exposures in PON1 subgroups and explored the possible effects of GW exposure combinations in relation to PON1 status.

 

[Hutan]

Methods
398 veterans from three pre-existing cohorts. 295 GWI cases and 103 controls

GWI cases were required to have multiple and/or moderate to severe symptoms that had persisted or recurred over six months in at least three of six defined symptom domains: (1) fatigue/sleep problems, (2) pain symptoms, (3) neurological/cognitive/mood symptoms, (4) gastrointestinal symptoms, (5) respiratory symptoms, and (6) dermatological symptoms. Veterans with diagnosed conditions that could account for their chronic symptoms or interfere with their ability to accurately report them (e.g., severe psychiatric disorders) were excluded as GWI cases.

Veterans with insufficient symptoms to meet the GWI case criteria and who reported no exclusionary medical or psychiatric diagnoses were classified as controls.

That sentence is slightly worrying. Hopefully the Gulf War veteran controls were actually healthy and not in some no mans land of being ill but not quite ticking the right boxes to be considered to have GWI.

One cohort was only asked to estimate their exposures during the war 9 to 15 years after the war, so that data may not be great.

The initial multivariable assessments of the association of GWI with GW exposures indicated that significant risk factors were limited to exposures that potentially affected acetylcholine levels or transmission.

That's important, although they haven't yet said what risk factors they considered.

Therefore, more detailed assessments of PON1 status in relation to GW exposures focused on exposures with a known or potential effect on acetyl- choline or acetylcholinesterase. For the purposes of the study, we considered all variables relevant to wartime exposure to pesticides, PB use, hearing chemical alarms (representing possible nerve agent exposures), and smoking during deployment as GW “cholinergic” exposures. Although some of these exposures (e.g., N,N-diethyl-meta-toluamide [DEET], PB, and some pesticides) are not known PON1 substrates, we included them based on their potential to act as AChEis and potentially affect acetylcholine levels or PON1 activity.

 

[Hutan]

PON1 status assays

The PON1 status assays were performed on plasma as previously described [35] be- tween 2004 and 2006 (SF-DOD cohort) and 2017 and 2020 (SF-VA and GWIC cohorts).

PON1 enzyme activity was evaluated in three substrates: paraoxon (to determine enzyme activity was evaluated in three substrates: paraoxon (to determine paraoxonase activity), phenyl acetate (for arylesterase activity), and diazoxon (for diazox-activity)

I think they must have identified the actual PON1 genotype, but I'm not 100% sure. Ed. I don't think they determined PON1 genotype - reference 35 only talks about determining PON1 function using substrates. They definitely determined the function i.e. the enzyme activity in breaking down the toxins. That does mean that the different ages of some of the blood samples might be relevant.

To determine PON1 status, data from the participants were sep-arated into three “functional genotype” PON1 activity groups (i.e., QQ, QR, and RR) by plotting the rates of diazoxon hydrolysis (diazoxonase [DZOase]) versus paraoxon hydrol-ysis (POase) (i.e., DZOase/POase ratio) [29] (see Figure 1). The separation of the three separation of the three PON1 activity groups was enhanced by carrying the assays out at a high salt concentration. Because the PON1R192 alloform is more sensitive to inhibition by high salt levels than
the PON1Q192 alloform, the use of high salt levels allows for a clearer separation of the functional genotypes.

 

[Hutan]

Results
Demographics
Females accounted for 18% of GWI cases and 15% of veteran controls. I don't know if the selection of the cohorts was on a population basis or if there was some selection

The cases and controls did not differ significantly by sex, race, ethnicity, military branch, or current smoking status. However, the GWI cases were younger, had fewer years of formal education, and included a higher proportion of enlisted personnel (vs. officers) compared to the controls

PON1 status distribution was nearly identical in GWI cases and controls, and there were no case–control differences in PON1 activity in the three substrates (Table 2).

Overall, PON1 activity in each of the three substrates differed by PON1 status, as expected [21,29] (Supplemental Table S1). Specifically, RR veterans had higher paraoxonase activity than QQ and QR veterans, while QQ veterans had higher diazoxonase activity than QR and RR veterans. There were no significant PON1 activity differences by GWI case status.

If PON1 status is determined by PON1 function, then of course PON1 status is going to vary by PON1 function. They don't seem to have determined PON1 genotype. ?

I'm surprised about the finding that PON1 activity didn't vary by GWI case status, but Table 2 shows very clearly that it didn't.

For the GW veterans from the SF cohorts who had their PON1 status assayed twice 11–15 years apart, all 24 veterans’ PON1 status remained constant over time. In contrast, their PON1 activity levels decreased over time, consistent with most previous reports, which have generally (though not uniformly [43]) identified reduced PON1 activity levels with age [44–46] (Supplemental Table S2).

 

[Hutan]

3.4 Association of exposures with GWI case status
Table 4 shows that a range of exposures were associated with GWI case status.

E.g.
86% of GWI cases heard chemical alarms sound versus 64% of controls. Adjusted odds ratio of 1.7
83% of GWI cases took the pyridostigmine bromide versus 60% of controls. Adjusted odds ratio of 2.0
13% of GWI cases wore flea collars versus 1% of controls. Adjusted odds ratio of 9.8
73% of GWI cases used pesticide spray on skin versus 36% of controls. Adjusted odds ratio of 3.9

The initial analyses provided preliminary indications that GWI risk associated with two cholinergic exposures (nerve agents and skin pesticides) may differ depending on whether veterans used PB or smoked during deployment (Supplemental Tables S4 and S5). In the multivariable models, hearing chemical alarms (reflecting possible nerve agent exposure) and using skin pesticides were significantly associated with GWI among veterans who took PB pills but not among veterans who did not take PB (Supplemental Table S4). Similarly, GWI risk in relation to hearing chemical alarms and using skin pesticides were substantially greater among veterans who regularly smoked during deployment compared to veterans who did not smoke (Supplemental Table S5). GWI risk associated with other cholinergic exposures did not appear to differ with PB use or deployment smoking status.

So, they are suggesting some interactions of the exposure factors.

3.5 GWI Status and Cholinergic Exposures by PON1 status
Table 5 gives the results of correlations between combinations of exposures and PON1 status on GWI status.

Hearing chemical alarms and having RR PON1 status is reported as giving the highest odds ratio (8.60). Having taken pyridostigmine bromide dramatically increased the odds in this groups of GWI (19.0).

Using pesticides on skin gave significant odds ratios for people with QQ and QR PON1 status (3.3. and 4.2). Having taken pyridostigmine bromide changed the odds ratios, increasing them in the QQ veterans to 6.3.

 

[Hutan]

I'm sorry, I'm still confused how they determined PON1 status and what they mean by it. I think they just mean the functional phenotype.

Furthermore, because of protein truncation mutations, some individuals genotyped as PON1 QR may express enzymes with PON1 QQ or RR phenotypes because only one allele is producing active PON1 [50].

Discussion
But, the good news - it seems their findings do fit with what is known about the biology of the PON1 forms.

The current study assessed associations of GWI with a range of deployment-related exposures in subgroups of GW veterans with different PON1192 status (QQ, QR, and RR). The first main finding of the study is that hearing chemical alarms during the war, our proxy for possible exposure to nerve agents, was significantly associated with GWI risk for RR, but not QQ or QR veterans. It is notable that despite differences in methodological approaches, this finding is consistent with previous reports by Haley et al. [26,27] of significantly elevated rates of GWI/Haley syndromes in relation to nerve agent exposure among veteran carriers of the PON1192 R allele, particularly RR homozygotes.

That's something to hang onto. The combination of exposure to nerve gas and the RR genotype as a major risk factor for GWI seems to have been replicated - they suggest in 'multiple studies'.

Nevertheless, multiple studies of veterans, including this one, have identified a signifi- cantly increased GWI risk in connection with hearing chemical alarms and other indicators of nerve agent exposure among GW veterans with PON1 RR status or the RR or QR geno- type and in veterans with low PON1192 Q alloform activity [26–28]. This suggests that GW veteran carriers of the PON1192 R allele, particularly RR homozygotes, had greater vulnerability to the effects of nerve agents during the Gulf War. Conversely, nerve agent exposure appears not to have posed a risk for veterans with PON1 QQ status/genotype. Whether this is the result of a direct protective effect provided by the PON1192 Q alloform or an indirect effect associated with as-yet-unidentified mechanisms remains to be deter- mined. For example, there are other enzymes besides PON1 (e.g., carboxylesterase and butyrylcholinesterase) that can bind and/or react with OP compounds such as sarin and reduce their toxicity [55,59,60]. It will be important for future studies to consider the role of these stoichiometric scavengers in relation to GWI risk.

And there does seem to be a coherent story about the increased risk for people with the Q alloform carriers with respect to pesticides:

Beyond looking at PON1 status/genotype and nerve agent exposure, we expanded our evaluation to assess the effects of other GW-related exposures in relation to PON1 status. The second main finding of this study is that using skin pesticides during deployment was significantly associated with GWI risk for QQ and QR status veterans but not for RR veterans. This pattern is consistent with reports that the PON1192 Q alloform is less efficient than the PON1192 R alloform in protecting against pesticide oxon forms such as chlorpyrifos oxon [61–64]. We know from government reports that chlorpyrifos and multiple other pesticides were used extensively during the GW [4].

The U.S. Department of Defense has reported that U.S. service members used at least 64 pesticide products during the GW and identified 15 pesticides of potential concern [8]. These included multiple organophosphates (e.g., chlorpyrifos, diazinon, malathion), car- bamates (e.g., methomyl, bendiocarb), pyrethroids (e.g., permethrin, d-phenothrin), and the organochlorine delouser lindane. Pesticides and repellants were often used in multiple combinations and for extended periods during the GW [4,8]. The skin pesticide most frequently used by GW veterans was DEET, including a high-concentrate (75%) form no longer used by the military [4,8].

 

[Hutan]

This was interesting, for beyond the Gulf War:

There is also suggestive evidence that individuals with the PON1 QQ/QR genotype are more susceptible to developing chronic symptoms in connection with chronic pesticide exposure compared to RR individuals. For example, one study found that QQ/QR farm workers were nearly three times more likely to have symptoms related to chronic pesticide exposure than RR farm workers [64].

They talk about how smoking can reduce PON1 efficiency and how that could explain some modification of risk associated with being a smoker.

They also talk about how the pyridostigmine bromide may have increase vulnerability for both RR veterans and QQ veterans.

Our exploratory findings indicate that the use of PB appears to have substantially increased GWI risk associated with hearing chemical alarms among RR veterans and GWI risk associated with skin pesticide use among QQ veterans. Both observations are consistent with a possible mechanism involving a reduction in PON1 activity with PB use that plausibly increased vulnerability to chemical nerve agents among RR veterans and vulnerability to pesticide use in QQ veterans.

Although PB was ordered for use during the Gulf War as a protective measure against potential deadly effects of nerve agent exposure, our findings support earlier indications that PB use may have had unintended consequences.

Although preliminary, these findings raise important hypotheses that require further testing in larger studies of GW veterans that are adequately powered to provide a detailed assessment of interactive effects of Gulf War cholinergic exposures in PON1 status subgroups.

It seems likely that a subset of people diagnosed with ME/CFS, people who aren't Gulf War veterans, have had cholinergic exposures that have caused their symptoms.