A Unique Circular RNA Expression Pattern in the Peripheral Blood of ME/CFS Patients, 2023, Cheng et al

[EndME]

ME/CFS is a debilitating disease with obscure aetiology. However, despite the extensive amount of research that has been performed on the transcriptomes of ME/CFS patients, all of them are solely focused on linear RNAs, and the profiling of circRNAs in ME/CFS has been completely omitted. In this study, we investigated the expression profiles of circRNAs, comparing ME/CFS patients and controls before and after two sessions of cardiopulmonary exercise longitudinally.

In patients with ME/CFS, the number of detected circRNAs was higher compared to healthy controls, indicating potential differences in circRNA expression associated with the disease. Additionally, healthy controls showed an increase in the number of circRNAs following exercise testing, while no similar pattern was evident in ME/CFS patients, further highlighting physiological differences between the two groups. A lack of correlation was observed between differentially expressed circRNAs and their corresponding coding genes in terms of expression and function, suggesting the potential of circRNAs as independent biomarkers in ME/CFS. Specifically, 14 circRNAs were highly expressed in ME/CFS patients but absent in controls throughout the exercise study, indicating a unique molecular signature specific to ME/CFS patients and providing potential diagnostic biomarkers for the disease. Significant enrichment of protein and gene regulative pathways were detected in relation to five of these 14 circRNAs based on their predicted miRNA target genes. Overall, this is the first study to describe the circRNA expression profile in peripheral blood of ME/CFS patients, providing valuable insights into the molecular mechanisms underlying the disease.

https://www.sciencedirect.com/science/article/pii/S0378111923004092

Highlights:

• Human peripheral blood circRNA profile changes in response to exercise.
• Distinct circRNA expression profile found in ME/CFS patients.
• CircLINS1 was differentially expressed on multiple days of exercise study.
• CircNUP98, circNPAT, circCSF3R and circLIN54 may serve as potential biomarkers.

The data of this study is taken from Bouquet et al. (2019) and was downloaded from the European Nucleotide Archive (ENA) using the study accession number PRJNA526259 in FASTQ file format (https://www.ebi.ac.uk/ena/browser/view/PRJNA526259).

 

[EndME]

This study being based on the study by Bouquet et al. means that the sample sizes are very small (ME/CFS (according to CCC) n=14, HC n=11). However, a re-analysis of other studies based on similar methods should be possible.

 

[Creekside]

Again, they're comparing inactive people to active healthy controls, so it doesn't necessarily mean much about ME.

 

[MSEsperanza]

The data of this study is taken from Bouquet et al. (2019)

Forum thread here.

Again, they're comparing inactive people to active healthy controls,

Controls were sedentary -- but agree that additional controls with other illness would be good. And a larger sample size.

 

[Creekside]

Sedentary is better than normally active, but not by much. A useful project for the OMF might be to establish criteria for suitable controls for ME research. Maybe a combination of different diseases (of specified severity), daily activity levels, and activity potential (how long can you ride an exercise bike, or do push-ups, or whatever?).

 

[Trish]

how long can you ride an exercise bike

Participants all did 2 day CPET as part of the study.

 

[Hutan]

How interesting. I'm excited to read the study.
If you, like me, don't know anything about circRNA other than it exists, you might find this paper helpful to read before tackling the ME/CFS study:
Circular RNA: metabolism, functions and interactions with proteins, 2020

Circular RNAs (CircRNAs) are single-stranded, covalently closed RNA molecules that are ubiquitous across species ranging from viruses to mammals. Important advances have been made in the biogenesis, regulation, localization, degradation and modification of circRNAs. CircRNAs exert biological functions by acting as transcriptional regulators, microRNA (miR) sponges and protein templates. Moreover, emerging evidence has revealed that a group of circRNAs can serve as protein decoys, scaffolds and recruiters. However, the existing research on circRNA-protein interactions is quite limited. Hence, in this review, we briefly summarize recent progress in the metabolism and functions of circRNAs and elaborately discuss the patterns of circRNA-protein interactions, including altering interactions between proteins, tethering or sequestering proteins, recruiting proteins to chromatin, forming circRNA-protein-mRNA ternary complexes and translocating or redistributing proteins. Many discoveries have revealed that circRNAs have unique expression signatures and play crucial roles in a variety of diseases, enabling them to potentially act as diagnostic biomarkers and therapeutic targets. This review systematically evaluates the roles and mechanisms of circRNAs, with the hope of advancing translational medicine involving circRNAs.

 

[Hutan]

This is background from the 2020 review paper I linked to:

CircRNAs formed from exons are generally localized to the cytoplasm [41]. ... Additionally, scientists have identified certain intron-containing circRNAs that are retained in the nucleus and regulate their parental gene expression [44, 45]. Some exonic circRNAs are predominantly distributed in the nucleus as well and increase the nuclear retention of proteins [46] or recruit proteins to chromatin [47]. Moreover, circRNAs can be delivered by extracellular vesicle (EV) and detected in the circulation and urine [48]. The sorting of these exosomal circRNAs seems to be regulated by associated miR levels in producer cells, while the specific biological activities transferred to recipient cells are largely unknown in diverse settings [48, 49]. The packaging, delivery and absorption of them also remain elusive so far. Recently, a handful of studies identified mitochondria-located circRNAs and examined their functions, which broaden our knowledge of circRNA derivation and mitochondrial transcriptome [50,51,52].

So these circRNAs can be inside cells, in the cytoplasm, in the nucleus and even in mitochondria. They can be transported between cells in extracellular vesicles, and are in the blood and in urine. Therefore, there are many places to look for the circRNAs. This 2023 ME/CFS study looks at the circRNAs in blood. That raises the question of whether a circRNA found in the blood is being excreted from a cell as a waste product, or if it is being transported for a purpose. It sounds as though there is a lot to find out.

circRNAs are relatively stable. It seems that a viral infection of a cell causes an up regulation of RNAase, which destroys the circRNAs. It's reported that reduced circRNAs can lead to 'aberrant PKR activation and autoimmunity'.

 

[Hutan]

The 2023 ME/CFS study has been done by a University of New South Wales team (Sydney, Australia). The supervising author (Michael Janitz) has a German connection also.

Sadly, this paper isn't open access. The snippets suggest that the team may be new to ME/CFS and have some ideas about it that are slightly odd. For example,

There is limited understanding on the aetiology of ME/CFS. Its symptoms are strongly associated with post-exertional malaise (PEM), which describes severe, prolonged fatigue after physical or mental labour (Brurberg et al., 2014).

In their review article, Rivera et al. (2019) proposed a hypothesis on the pathogenesis of ME/CFS, labelling it “Three Pillars” based on these three biological systems. The theory proposes that the primary tissues that are involved in the development of the illness are the neurons and glial cells in the CNS, lymphocytes Th1/Th2, NK cells and B cells in the immune system, as well as the hypothalamus–pituitary–adrenal axis in the endocrine system (Rivera et al., 2019).

Currently, ME/CFS has diagnostic criteria based on clinical features and its diagnosis relies on the exclusion of other medical conditions (Fukuda et al., 1994, Carruthers et al., 2011).

Autoantibody and oxidative stress are the most established types of biomarkers of ME/CFS, but their efficacies are deemed inconclusive (Hickie, 1999, Mitchell, 2016).

That's a surprising reference to use for PEM, and the definition isn't really right. And, it's the first I've heard of the Three Pillars idea - I don't think we've seen much evidence to support the HPA axis idea or the endocrine system being central in the pathogenesis of ME/CFS. And they are out of date with the criteria they cite. I don't think it's accurate to say that there are autoantibody and oxidative stress biomarkers. These things are not a big problem if the authors are experts in circRNA. It just suggests that they aren't very well informed about ME/CFS, maybe they just put in what came up when they googled, and so they might possibly put an unusual interpretation on any circRNA findings.

There are additional factors that make the diagnosis of ME/CFS difficult, including the underestimation of the severity of ME/CFS and the incorrect judgement of its nature to be psychogenic rather than physiological, due to the inexperience of some health service providers (Clayton, 2015).

They do dismiss the idea of ME/CFS being psychogenic, so it looks as though they mean well.

 

[MSEsperanza]

Sadly, this paper isn't open access.

Elsevier journals -- Access for healthcare and patients

Patient access
Patients and caregivers can make requests for individual papers related to medicine and healthcare at no cost. These papers can be requested by e-mailing patientaccess@elsevier.com.

Requested the full text of this paper yesterday and got it today.

 

[Trish]

The data of this study is taken from Bouquet et al. (2019) and was downloaded from the European Nucleotide Archive (ENA) using the study accession number PRJNA526259 in FASTQ file format (https://www.ebi.ac.uk/ena/browser/view/PRJNA526259).

The snippets suggest that the team may be new to ME/CFS and have some ideas about it that are slightly odd.

Fortunately the people who did the original research and provided the data this was based on do, I assume, understand ME/CFS and how to diagnose it, since several of the Workwell team including Staci Stevens and Mark VanNess are co authors of the original data, so I assume they diagnosed the patients and did the 2 day CPET.

The patients in the original study were all female, so this data comes from an all female sample. Given differences being found in other studies between male and female biomedical results following CPET, it would be good if any replication studies used both male and female patients and included separate analyses of results.

 

[SNT Gatchaman]

Sadly, this paper isn't open access. The snippets suggest that the team may be new to ME/CFS and have some ideas about it that are slightly odd.

I thought the same as I read through. I ended up with quite a few highlighted sections I thought interesting.

Complementing what @Hutan said above —

CircRNAs are a type of non-coding RNA generated by the backsplicing of precursor mRNA transcripts to produce exonic circRNAs, and to lesser extent, intronic RNAs [...] The backsplicing process results in a ‘looped’ RNA structure which covalently joins the 3' end of a downstream region to the 5' end of an upstream region. The ligation interphase produced from this is known as the backsplice junction (BSJ)

Joining the 5' and 3' ends to form a loop or circular form means that those ends are unavailable to promote degradation, so they survive longer in plasma —

CircRNAs exhibit appealing biomarker properties such as stability and ubiquitous expression across human tissues [...] exceptional resistance to exonucleases due to their circular structure which lacks 3’ and 5’ ends. This feature grants circRNAs remarkably longer half-lives than linear RNAs—an average of 48 hours compared to 10 hours

ME patients have more circRNAs (including uniquely expressed). Controls showed an increase following the exercise challenge, while levels in ME patients remained statically elevated —

We identified that the number of non-redundant circRNAs in ME/CFS patients was higher compared to healthy ‘control’ individuals across all time points over the 7-day period. We also observed a trend of increasing circRNAs in controls as days progressed, while the numbers in ME/CFS patients exhibited no trend across the entire time frame.

A visualisation of significant DE [differentially expressed] circRNA transcript distribution in the form of volcano plots shows that the largest number of DE circRNAs was detected on Day 2 (19), after one session of CPET. Interestingly, the second highest number of DE circRNAs was observed on Day 7 (9), five days after the second session of CPET. [...] may be caused by perturbation in the circRNA transcriptome after exercising, combined with the possibility that the circRNA transcriptome is expressed and regulated differently between ME/CFS patients and healthy individuals.

They used the term "uniquely expressed" as a subset of "differentially expressed" when a circRNA was only seen at all in one condition (ME or HC) rather than present in both but at higher levels in one —

We also considered circRNAs that were expressed in only one condition as DE, and we named them ‘uniquely expressed’ circRNAs to distinguish them from significantly DE circRNAs.

HCs increased the numbers of uniquely expressed circRNAs in the days following first CPET. However ME patients consistently had more uniquely expressed circRNAs.

There was an increasing number of uniquely expressed circRNAs in control samples from Day 1 to Day 7 while the number of uniquely expressed circRNAs in ME/CFS remained at similar levels. The number of uniquely expressed circRNAs in ME/CFS patients, ranging from 1472 to 1757 across all days, was much higher than the number of uniquely expressed circRNAs in control samples

 

[SNT Gatchaman]

Day 1 baseline (2 hours pre-CPET #1) at top left shows ME overexpress circRELL1.

On Day 1, there was only one significantly DE circRNA, circRELL1. There is no publication on circRELL1 related to neurological disorders. However [...] circRELL1 has been suggested to indirectly regulate autophagy activation by acting as a sponge for miR-637, a miRNA that suppresses the transcription of NUPR1 in the autophagy pathway. CircRELL1’s ability to regulate autophagy is of particular interest due to its association with the immune system aspect of ME/CFS pathology.

Referencing Exosomal circRELL1 serves as a miR-637 sponge to modulate gastric cancer progression via regulating autophagy activation (2022, Nature Cell Death & Disease)

 

[SNT Gatchaman]

The circTFRC in this study was identified to be expressed uniquely in ME/CFS patients, with particularly high expression on Day 1 and Day 7, when no CPETs were conducted by the subjects. It was not expressed at all on Day 2 and Day 3, when subjects had undergone CPETs the previous day. This suggests that this circTFRC’s production is affected by exercise and is differentially expressed in ME/CFS patients compared to healthy individuals, implicating its prospect as a diagnostic biomarker for ME/CFS. circTFRC might be involved in the regulation of genes controlling cell proliferation and differentiation, as well as the immune response. These findings suggest that circTFRC may play a role in coordinating the cellular response to various physiological stressors.

Referencing The biogenesis and biological functions of circular RNAs and their molecular diagnostic values in cancers (2019, Journal of Clinical Laboratory Analysis)

 

[Hoopoe]

On the function of NUPR1

Transcription regulator that converts stress signals into a program of gene expression that empowers cells with resistance to the stress induced by a change in their microenvironment

This suggests circRELL1 is an indicator that cells are undergoing significant stress

Edit: or maybe that sensitivity to stress signals is being increased a lot.

 

[SNT Gatchaman]

We identified that the circATP2B4 isoform [...] expressed by the ATPase plasma membrane Ca2+ transporting 4 gene (ATP2B4) was the sixth most uniquely expressed circRNA on Day 1, as well as the second most significantly DE circRNA on Day 7. On both days, this circATP2B4 had a higher expression level in ME/CFS patients than in controls.

It is possible that circATP2B4 is usually expressed at a higher level in ME/CFS patients than in healthy individuals before any exercise or a long time after the last exercise session, but this higher expression becomes diminutive after immediate exercise. The exact function of circATP2B4 is still being investigated, but early studies have suggested that it may play a role in regulating the expression of ATP2B4, which encodes an ion transport enzyme that is crucial in maintaining intracellular calcium homeostasis. This channel is important for maintenance of cellular homeostasis, including the regulation of blood pressure, heart rate, and insulin secretion

However, as most circRNAs do not translate into proteins and have different molecular functions to linear RNAs, there is insufficient evidence to suggest that circATP2B4 acts on the same biological pathways as its corresponding mRNA transcript.

 

[Grigor]

I asked associate professor Martijn Luijsterburg who's specialized in DNA repair about the study. This paper is way above my head really, but he was kind enough to answer me anyways.

 

[Hutan]

Method

On Day 1 and Day 2, when CPETs were conducted, whole blood samples were taken two hours prior to the session of CPET. Whole blood samples were also taken on Day 3 and Day 7 via follow-up visits.

I haven't got to the end of the paper yet.

I think the idea of this paper is really good. We now need it to be replicated in much larger samples, and tied to detailed data on activity levels and symptoms. I think we need samples to be taken periodically from each person and correlated with activity levels etc. CircRNAs are found in saliva too - looking in saliva might make multiple sampling easier.

I expect the next few years are going to see major leaps in the understanding of what the circRNAs do, and that will help us.

Crossposted with Grigor - Martijn sounds like a handy friend to have.

 

[Creekside]

Participants all did 2 day CPET as part of the study.

But did whatever was measured in the CPET result in controls being rejected? What I was suggesting is a standard criteria for controls for ME research. Hmmm, even better: a carefully selected group of controls who are subjected to different factors (exertion, maybe injected with IL-6 or LPS, more sleep and less sleep, etc) whose data is available for any research projects. If you want to compare ME patients with controls for circular RNA levels after exercise, match your subjects exercise tests with what is available from the controls. If your research involves some factor that the control group hasn't provided data on yet, you can request they be tested for that (and compensated, of course).

This would be a big change (I think?) in how research is done, but it would save time and money and improve the quality of the results, since the control data would be consistent. Surely this is done in other sciences, such as climatology, astronomy, or seismology, using historical data.

 

[Hutan]

If you want to compare ME patients with controls for circular RNA levels after exercise, match your subjects exercise tests with what is available from the controls.

A problem is that it isn't that straightforward. People with ME/CFS have highly variable results on CPETs - that seems to be almost a diagnostic feature. When I did two CPETS, the first one showed that I had normal fitness, the second one showed that my fitness was well below normal.

So, which CPET would you choose to match controls on? If it's the first one, can you be sure that any activity in advance of the CPET, perhaps even just preparing to get to the clinic, hasn't affected the person with ME/CFS? For example, in the two days prior to my first CPET, I had prepared the house so that my son could cope for a few days on his own, shopping, filling the fridge with prepared food, I packed, stood in airport queues, flew to another city, chatted to the taxi driver, and settled into my accommodation. On the day of my first test, I stood by the roadside in the cold for 15 minutes waiting to be picked up. If things aren't explicitly controlled, there will be confounders.

Probably the best matching for fitness would be on something like step counts over a sufficiently long period and with clear directions about pre-test activity and diet. Probably you need to have people wear activity monitors in the days before, during and after the test to ensure compliance.

This would be a big change (I think?) in how research is done, but it would save time and money and improve the quality of the results, since the control data would be consistent. Surely this is done in other sciences, such as climatology, astronomy, or seismology, using historical data.

I agree there could be more use of standard control data, it would reduce costs and make research quicker. But protocols would have to be very specific and standardised. And I think there's probably always a need for some controls samples to be processed along with the patient data, to make sure the researchers haven't inadvertently introduced some factor that changes the results. I think ideally you'd have a matched control sample of 30 or so that is processed along side the patient sample, but you would also report against results for larger samples of healthy controls and disease controls. I don't think that's very different to what researchers try to do now; probably there is just a lack of comprehensive comparison data collected with standard sampling protocols right now.

I agree though that research funders should give more funding for producing benchmark metabolomic and proteomic data for healthy controls covering different ethnicities, ages, fitness levels and sex, including keeping anonymised data in public databases so that subsamples can be selected for tighter comparisons with disease groups.