Preprint Flow Clotometry: Measuring Amyloid Microclots in ME/CFS, Long COVID, and Healthy Samples with Imaging Flow Cytometry, 2024, Pretorius

[Dolphin]

https://www.researchsquare.com/article/rs-4507472/v1

Article

Flow Clotometry: Measuring Amyloid Microclots in ME/CFS, Long COVID, and Healthy Samples with Imaging Flow Cytometry

Etheresia Pretorius Stellenbosch University Massimo Nunes Stellenbosch University Jan pretorius Douglas Kell University of Liverpool

https://doi.org/10.21203/rs.3.rs-4507472/v1



Myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) has received more attention since the characterization of Long COVID (LC), a condition somewhat similar in symptom presentation and, to some extent, pathophysiological mechanisms.

A prominent feature of LC pathology is amyloid, fibrinolysis-resistant fibrin(ogen) fragments, termed microclots.

Despite prior identification of microclots in ME/CFS, quantitative analysis has remained challenging due to the reliance on representative micrographs and software processing for estimations.

Addressing this gap, the present study uses a cell-free imaging flow cytometry approach, optimized for the quantitative analysis of Thioflavin T-stained microclots, to precisely measure microclot concentration and size distribution across ME/CFS, LC, and healthy cohorts.

We refer to our cell-free flow cytometry technique for detecting microclots as 'flow clotometry'.

We demonstrate significant microclot prevalence in ME/CFS and LC, with LC patients exhibiting the highest concentration (18- and 3-fold greater than the healthy and ME/CFS groups, respectively).

This finding underscores a common pathology across both conditions, emphasizing a dysregulated coagulation system.

Moreover, relating to microclot size distribution, the ME/CFS group exhibited a significantly higher prevalence across all area ranges when compared to the controls, but demonstrated a significant difference for only a single area range when compared to the LC group.

This suggests a partially overlapping microclot profile in ME/CFS relative to LC, despite the overall higher concentration in the latter.

The present study paves the way for prospective clinical application that aims to efficiently detect, measure and treat microclots.

Health sciences/Medical research/Biomarkers/Diagnostic markers

Biological sciences/Biological techniques/Imaging/Fluorescence imaging


Expanded Key Points for Review Only

• We have recently developed a method to detect microclots in plasma samples using cell-free flow cytometry analysis, which we have coined 'flow clotometry'.
• Quantitative data presented in this paper corroborate that microclots are present in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) plasma samples at significant levels when compared to ‘healthy’ participants, and that Long COVID patients exhibit the highest concentrations of microclots, with an 18- and 3-fold change in their median compared to the healthy participants and ME/CFS group, respectively.
• The size distribution of microclots in the ME/CFS group is significantly different from that of healthy participants in all area ranges, but were only significantly different in one area range when compared to the LC group. Hence, the size distribution of microclots in ME/CFS samples is broadly comparable to that of LC samples.
• Importantly, microclots are present in an easily accessible and measurable fraction of blood, and hence the implementation of flow cytometry (clotometry) microclot assessment in the clinical setting is warranted. Such methods can provide a high throughput, and deliver quantitative information regarding both burden and size distribution of fibrinaloid microclots.

Main Key Points

• We have developed a method to detect microclots in plasma samples using cell-free flow cytometry analysis, which has been coined 'flow clotometry'.
• Quantitative data corroborate that microclots are present in myalgic encephalomyelitis/chronic fatigue samples at significant levels when compared to healthy participants

https://twitter.com/user/status/1807033288864698458

https://twitter.com/user/status/1807034666471350772

https://twitter.com/user/status/1807036308365144570

 

[John Mac]

Does anyone know if this group have used healthy controls in their previous studies?

ETA: I only ask as if they haven't this could be the first evidence that the blood clots seen in the previous studies were not an artefact of the testing process and were genuinely in the patients blood. That is of course if the healthy control samples were collected/stored/processed/tested exactly the same as the patient samples.

 

[SNT Gatchaman]

Does anyone know if this group have used healthy controls in their previous studies?

Yes. See figures in this post from Hutan in Oct 2021, which is for the original LC microclot paper. That paper had followed their acute Covid microclot paper, which itself followed the pre-Covid evaluation of diabetes. All these amyloid fibrin papers followed their original observation of "dense matted deposits" via electron microscopy, before they tagged them with Thioflavin T to indicate they were amyloid.

The technique used in this pre-print also uses Thioflavin T, but the advance seems to be quantitive assessment by object count (of specific size range), versus the overall area of green signal by image analysis previously. Still not confirming they are circulating in vivo. Perhaps the fibrin PET studies will help answer that question.

They comment in discussion —

Previously, our microscopy analysis and subsequent ImageJ assessment of micrographs led to the inference that microclots are present in our ME/CFS cohort at a level 10x that of the healthy participants group. In the present study, cell-free flow clotometry analysis of ME/CFS PPP samples incubated with ThT reflect the same direction of significance but a slightly lower fold change in the concentration of microclots: in terms of objects/mL, the present experiment revealed that the plasma of the ME/CFS cohort contains over 5x the number of microclots than those of the healthy participants (Table 2 & Fig. 1A) (***). The present flow clotometry analysis corroborates quantitatively that ME/CFS individuals contain significantly higher levels of microclots in their circulation than do healthy participants, albeit less than inferred from our previous microscopy analysis. With that being said, our previous study assessed area (% area amyloid) across micrographs and therefore cannot be reliably compared to the present quantitative technique where microclot concentration (count) was measured.

Another important parameter assessed in this study is the prevalence of microclots within defined area ranges (Fig. 2). In the present study, the ME/CFS group measured with a higher count within all area ranges when compared to healthy participants, of which the greatest differences were recorded in the 100–400µm2 (***) and 400–900µm2 (***) ranges. Ultimately, we provide evidence that both the concentration and prevalence of large microclots in ME/CFS PPP samples are significantly greater when compared to a healthy participant cohort. When assessing the percentages of distribution across area ranges, it is apparent that the ME/CFS group has an overall greater prevalence of larger microclots than does the LC group

 

[Murph]

Strong paper. It provides further evidence for a fascinating observation in the disease; an observation that could suggest interventions. I'm pleased this avenue of enquiry is making progress.

 

[Jonathan Edwards]

I have not read the paper but there still seems to be no acknowledgement that particles of this size, if present in vivo, would have been removed by centrifugation? The sizes they quote are about monocyte size and would be removed by a brief low speed spin - much faster than red cells would.

That appears to mean that these particles develop newly in the plasma when incubated with thioflavine T. They do not appear to be clots in any normal sense, in that they are not composed of an extended fibrin lattice but more like a precipitate of protein of the sort seen when you add trichloroacetate or heat plasma in the Bence-Jones test. It is unclear why this sort of precipitation should occur in vivo and why it should occur at such a chronic low level that no pathology is detectable. Microparticles in venous blood of this sort would be expected chiefly to produce a pneumonitis and ultimately pulmonary hypertension as in pulmonary veno-occlusive disease.

 

[Hutan]

We have previously shown via fluorescence microscopy that ME/CFS plasma samples contain significant levels of microclots 13. While the utmost care was taken in our previous study to ensure representative micrographs were obtained for each sample, it is almost impossible to determine the exact concentration of sample constituents (in this case, microclots) by microscopy (qualitative) analysis unless one is able to analyse an entire sample 15,17,35,36. We have recently developed a solution to this issue: cell-free flow cytometry analysis of microclots in plasma samples 35, which, in actuality, is not true flow cytometry – hence the term flow clotometry. The assessment of plasma via this method offers robust and more reliable measurements of microclots – furthermore, we possess an imaging flow cytometer which allows the visualization of each event, thereby allowing both quantitative and qualitative analysis of microclots, and other constituents.

It does sound like an advance, a system for measuring these microclots in a more objective way.

Stored PPP samples from 16 healthy individuals (10 females; 6 males), 30 individuals with ME/CFS (22 females; 8 males), and 30 individuals with LC (17 females; 13 males) were used for this study.

The ME/CFS population was previously recruited from the ME/CFS Foundation of South Africa and blood collection was performed at PathCare. This cohort includes 25 samples that were analyzed in a previous study (Nunes et al., 2022). All ME/CFS participants in this study had not experienced a SARS-CoV-2 infection prior to the date of sample collection.

LC individuals were recruited from our clinical collaborator's practice.

Sodium citrate tubes were used for blood collection, and centrifuged at 3000×g for 15 min at room temperature. The PPP was collected, and stored at -80°C.

After samples were left to thaw at room temperature, 47µL of PPP was incubated with 3µL of thioflavin T (ThT) (Sigma-Aldrich, St. Louis, MO, USA) (final exposure concentration of 0.03mM) for 30 minutes at room temperature in a dark room. The samples were then transferred to the Amnis® FlowSight® Imaging Flow Cytometer (Luminex).

 

[Trish]

If the microclots are, as suggested, an artefact of the process and only present in vitro, then the question that needs to be asked is, why are there more microclots formed in plasma of pwLC, an intermediate number in pwME and the least in healthy controls. Is there some factor in the blood that is more/less present in sick people than in healthy people that is induced by the processing to form microclots.

 

[Hutan]

Table 1 says that all samples were previously stored platelet poor plasma. One thing I wonder about is the storage conditions - length of time, maybe some differences in the storage protocols between the groups?

The fact that a significant number of the ME/CFS samples had been analysed for microclots in a previous study is obviously less than ideal, creating the possibility that samples with higher microclot densities were selected for inclusion. Not that that is necessarily what happened of course.

 

[Hutan]

In addition, the difference in data between fresh and stored samples still requires additional investigation, and holds relevance to the clinical implementation of this cell-free flow clotometry technique. In this study, we used stored plasma samples and not freshly obtained blood samples. We are currently conducting such an experiment where we are investigating whether the microclot concentrations and sizes change as a result of freezing (and thawing).

Ah, excellent, they are aware of that problem and looking into at least one aspect of the storage/processing questions.

 

[Jonathan Edwards]

Is there some factor in the blood that is more/less present in sick people than in healthy people that is induced by the processing to form microclots.

This is the interesting question. It might be that a shift in concentration or activation of amyloid-forming proteins occurs in LC and ME/CFS and that this assay is a way to show that, but nothing to do with circulating micro clots in people blocking vessels.

The use of the previous ME/CFS cohort is way less than ideal. If I remember rightly this population had a very atypical mix of other comorbid conditions including several with rheumatoid arthritis. It makes a mockery of the whole thing if they are re-measuring these. These samples will also be very old now and freezing plasma might do all sorts of things to the formation of precipitates.

 

[Hutan]

If I remember rightly this population had a very atypical mix of other comorbid conditions including several with rheumatoid arthritis.

Yes, see Table 1.
e.g. 10% of the ME/CFS sample had rheumatoid arthritis; 17% had psoriasis.

 

[EndME]

Ah, excellent, they are aware of that problem and looking into at least one aspect of the storage/processing questions.

I think they should be more than just aware of this. If I recall the results by Dalton et al, they had shown that this kind of storage completely nullifies any results and makes them useless. Haven't gone through the study yet, but the deciding factor would possibly whether controls had been exposed to Covid which seemed to be a driving force in the results of Dalton et al.

 

[Trish]

It seems like the sort of situation where a next step might usefully be to take fresh blood samples from a restricted population group to enable easy comparison.

For example, women with normal BMI, not on any medication, with no co-morbidities, and pre menopause, and as a pilot study take ten each of pwME, pwLC and sedentary healthy controls, and possibly another disease group where Pretorius says they have found microclots.

Then ensure the scientists are blinded to the group for each participant, and see what results.

 

[SNT Gatchaman]

there still seems to be no acknowledgement that particles of this size, if present in vivo, would have been removed by centrifugation? The sizes they quote are about monocyte size and would be removed by a brief low speed spin - much faster than red cells would.

If it's a density gradient and the clots are "wispy" then maybe they could stay above cells? But perhaps that's irrelevant at 3000g and most anything particulate should be at the bottom of the tube? Although in counter to that they always make the point that it's platelet-poor plasma not platelet-null plasma, so at least platelets can be in the supernatant, so maybe also things of similar density, esp given that the platelets spread out when activated.

 

[Jonathan Edwards]

If it's a density gradient and the clots are "wispy" then maybe they could stay above cells?

But they aren't wispy. We have seen them on immunofluorescence and they are lumps. They are solid protein, so a lot higher density than cells, and they are as big as monocytes. The only thing that would stop these falling to the bottom of the tube even without centrifugation is the red cells - which are removed. These things should sediment about a thousand times faster than a platelet, which is lipid rich.

 

[Jonathan Edwards]

It seems like the sort of situation where a next step might usefully be

Spot on @Trish

 

[Mij]

Table 1 says that all samples were previously stored platelet poor plasma. One thing I wonder about is the storage conditions - length of time, maybe some differences in the storage protocols between the groups?

A side note. My samples were sent overnight on dry ice and had to be tested within 24hrs.

 

[Hutan]

If you look at the charts in Dolphin's first post above, not all of the ME and LC samples had elevated levels of microclots. So, if elevated levels of microclots are a characteristic of particular disease states, then either not everyone in the samples has those diseases, and/or it's a characteristic that isn't present all of the time.

See this study:
The influence of environmental variables on platelet concentration in horse platelet-rich plasma, 2016, Rinnovati et al

It suggests that a range of environmental variables could influence what is in the blood samples during storage and in subsequently processed fluids - things like NSAID use, hydration, sample collection time of day, exercise, gender and age.