Prevotella & endemic TB might explain low COVID 19 deaths in India- Sandeep Chakraborty (Interview)
With experts trying to find the possible reasons for the low COVID related death rate in India, a hypothesis by Sandeep Chakraborty has generated considerable interest because of the data based explanation it offers. Sandeep Chakraborty (B-Tech, Electrical Engineering IIT Kharagpur, 1997), (Masters. Computer science, IIT Kharagpur, 2008) is a researcher focused on computational protein modelling, genome projects, and designing anti-microbial peptides. His hypothesis based on his current research and data analysis on COVID-19 pathogenesis suggests that it's Prevotella - the anaerobic bacteria that might be responsible for the systemic complications seen in COVID-19. Another component of his hypothesis depends on data from previous independent studies showing that aerobic Mycobacterium Tuberculosis limits the growth of anaerobic species Prevotella. Sandeep's hypothesis says this might be one of the main reasons for India’s apparent low numbers of severe cases/ deaths due to COVID19 in India - his exclusive interview with Dental Tribune South Asia.
1. Hi Sandeep, your hypothesis on Mycobacterium TB and Prevotella has generated massive interest and discussion on social media. How did you come up with this theory? How did it all start?
Well, a little bit of history. It was late Jan 2020, and I was on the lookout for sequencing data from COVID 19 patients. This is where I pick up all my sequencing data from https://www.ncbi.nlm.nih.gov/sra/
While analyzing the first paper from China  to report the SARS-CoV-2 genome I noticed a complete predominance of Prevotella (this was BALF sample). At least in this patient, this was a serious secondary infection . (BALF - Bronchoalveolar lavage fluid)
A few days later, I realized the Chinese paper had also noted this abundance in the Supplementary Tables, but made no mention of it in the paper. So, in that sense this is verified.
It was already in the back of mind, that tuberculosis might play a role - either positive or negative -in India. However, after the lockdown - and the passing of the average 2 weeks of incubation required for symptoms - when I saw that the numbers were low, and the hospitals were not really filling up, I started looking for papers on tuberculosis and Prevotella. That's when I stumbled on this hypothesis.
2. So what is your hypothesis exactly?
Metagenomic data of Covid19 patients from across the globe shows a certain distinct pattern- anaerobic bacteria (Prevotella primarily) seem to predominate, ie there is a disruption of balance (homeostasis) with aerobic counterparts. This does not happen in healthy patients. Other independent studies have shown that Mycobacterium Tuberculosis (aerobic bacteria) limits the growth of Prevotella. My hypothesis links these two facts - to postulate that this may be a reason for India’s low COVID-19 severe cases/ deaths. Note, that this does not speak about infections - but the observed limited effect of these infections till date.
3. So that means, the role of Prevotella is very significant in the pathogenesis of COVID-19. It also brings us to another important point - Prevotella is not new to us, so there must be some published reports on other systemic infections caused by Prevotella. Is there any medical condition that has COVID manifestations but known to be associated with Prevotella?
Yes, there is a condition - Lemierre’s syndrome.
Many CVID 19 symptoms - especially the serious ones like ARDS, blood clots and septic shock - are also found in a rare ”forgotten” disease, Lemierre syndrome, but with one key - and important - difference: Lemierre syndrome originates in the jugular vein, whereas Covid19 emanates from the lungs (which makes treatment much easier). In Lemierre’s syndrome, anaerobic bacteria (mostly Fusobacterium, but also Prevotella) colonize the jugular vein (happens mostly in children and young adults) through the peritonsillar blood vessels.
Here, forms a thrombus (blood clot) in which resides the bacteria, which can dislodge and travel through the bloodstream to other organs. They can then form abscesses and septic infarctions (obstruction of the blood supply). Organs like the lungs, liver, spleen, kidneys and nervous system are thus effected. Please go through these scientific references [1, 2].
Another interesting similitude is the key role of viruses - just as I postulate that SARS-COV 2 enables anaerobic bacteria, it is known that Epstein Barr virus (EBV) enables LS . Thus, co-infection of SARS-Cov2 and EBV is a distinct possibility, already observed in a few patients .
4. There are some reports of Disseminated Intravascular Coagulation observed in COVID cases. Any association with Prevotella?
As mentioned above, the thrombus (blood clot) is a key step in the genesis of the serious form of the disease. And Prevotella (also Fusobacterium and other bacteria) have the required enzymes to do this - specifically, hemolysins which break down red blood cells, and heme-binding proteins which can sequester the heme (which binds iron, which binds oxygen). This leads to the breathlessness observed in almost all COVID 19 patients. And once again, this is not a hypothesis - sequencing data from Covid19 patients show that these proteins are being expressed.
Finally, deep sequencing of 3 blood samples did not yield any virus in a Chinese study (where the BALF had significant viral load) . Thus, the theory that the virus attacks haemoglobin chain does not appear to be true. Again, this adds credence to the bacterial hypothesis - something is definitely breaking down haemoglobin, causing clots.
5. Do you have any medical research or microbiology/ virology background to come up with this? Your profile says you have done both - B-Tech and Masters from IIT- KGP.
No, I don't have a medical degree. I am self-taught in this aspect and can converse with a doctor reasonably well.
6. What components of this theory are already established pieces of evidence and what has yet to be established?
Well, I have established that SARS-CoV-2 enables Prevotella to colonize the lungs (it's yet to be peer-reviewed, but you have to take my word for it now). The other half of the theory - i.e. MTB doesn't allow Prevotella to flourish has been established. A recent study notes that `MTB+ patients were enriched with Anoxybacillus, while MTB- patients were enriched with Prevotella, Alloprevotella, Veillonella, and Gemella' . This observation is reiterated in another study - `yet in pulmonary tuberculosis, contrary to the expectation, Prevotella species are decreased rather than increased' , which goes ahead to propose a mechanism to this resistance - MTB exopolyphosphatase. In short, maybe MTB does not allow anaerobic bacteria to break the homeostasis in their favour.
7. How does your theory connect with Azithromycin - an antibiotic that has generated considerable interest?
Azithromycin fits in with the anaerobe colonization theory as an antibiotic. Assuming most MTB is resistant to Azithromycin, I would like to think it will help. But I think there are better Antibiotics for the anaerobic bacteria.
8. How does this whole M. tuberculosis thing connect with the BCG vaccine? And is this the reason why BCG vaccination is being tried as an option to COVID19?
There are theories that the BCG vaccine provided innate immune memory . It does have some legs.
9. Does someone need to have active M. tuberculosis in the body to prevent Prevotella?
I would think so - see Q6, you would need MTB exopolyphosphatase.
10. Not sure, but it's said that in India people have subclinical TB. What is your opinion on those people in India who have never had TB history? Will they have the same susceptibility to covid19 as western people?
I would also like to believe Indians would have better immunity, although I am not sure. Anyone who can have Pani-Puri at Lower Parel, and not head directly to the hospital, would have some kind of strong immunity.
11. If this theory gets validated, can we afford to breathe easy because then our target shifts from coronavirus to Prevotella?
Yes, I believe so - a multi-pronged attack (some anti-viral that works even a little) would be even better. I am not sure of HCQ, but I hear low dosage helps. I would like to believe - as reports from many parts of the world are saying - that it has already become more endemic than we know.
12. Why can't the intervention start early? What can be done as soon as the patient tests positive? How will your theory help in this regard?
It can. In fact, it could also act as a prophylactic in critical areas - one has to be careful about indiscriminate use to stop antibiotic resistance.
13. Could we extrapolate your findings to other regions? If yes, in Ecuador- where TB is a major issue but the death rate due to covid19 is also high.
Can't really say - would have to think, does high altitude have a role? I have analyzed one patient sample from Ecuador, and there the viral load was very low (and the bacterial homeostasis was not broken).
14. What about Australia New Zealand.. far fewer cases. TB isn't endemic there. They have low prevalence as well as the low death rate due to COVID-19. How would you explain that? In fact, they have managed to flatten the curve.
They have better health-care systems, so early detection (that matters) - and social distancing is easier. I have not seen their mortality rates.
15. What is the role of- virulent strains of virus or say basic demographics of the population? India has very low Chinese contingent. Europe and the USA have a huge Chinese contingent.
I think that would make a difference in the initial spread - since this coincided with the Chinese New Year and many people travelled. But, this virus seems to spread fast - and patients remain asymptomatic for 20+ days. So, in the long run, I don't think demographics would matter.
Dental Tribune South Asia acknowledges the contribution of Drs. Aashutosh Karnik (MDS, Periodontics) and Avadhoot Avadhani (PhD, Oral Biology) for adding their questions.
1. Garza-Alatorre A, Hern ́andez-Rosales C, Rodr ́ıguez-Coronado J, Sol ́ıs-Gonz ́alez MC, Balderrama-D ́avila R, et al. (2015) Atypical lemierre’s syndrome caused by prevotella oris. Medicina universitaria 17: 218–221.
2. Hadjinicolaou AV, Philippou Y (2015) Lemierre’s syndrome: a neglected disease with classical features. Case reports in medicine 2015.
3. Chacko E, Krilov L, Patten W, Lee P (2010) Lemierre’s and Lemierre’s-like syndromes in association with infectious mononucleosis. The Journal of Laryngology & Otology 124: 1257–1262.
4. Chen T, Song J, Liu H, Zheng H, Chen C (2020) Positive Epstein-Barr virus detection in coronavirus disease 2019 (COVID-19) patients. Available at SSRN 3555268.
5. Wu F, Zhao S, Yu B, Chen Y, Wang W, et al. (2020) Complete genome characterisation of a novel coronavirus associated with severe human respiratory disease in Wuhan. China bioRxiv 24.
6. Chakraborty S (2020). Sars-cov2 enables anaerobic bacteria (prevotella, et al) to colonize the lungs disrupting homeostasis - symptoms (ards, septic shock, blood clots, arterial stroke) finds resonance, with key differences, in the ‘forgotten disease’ Lemierre syndrome, caused by anaerobic bacteria enabled by Epstein Barr virus. doi:10.31219/osf.io/usztn. URL osf.io/usztn.
7. Hu Y, Kang Y, Liu X, Cheng M, Dong J, et al. (2020) Distinct lung microbial community states in patients with pulmonary tuberculosis. Science China Life Sciences.
8. Block J (2019) Exopolyphosphatase of mycobacterium tuberculosis might limit the growth of bacteria which thrive in the inflamed and injured lung. Arch Microbiol 3: 102–106.
9. Kurthkoti K, Das G (2020) Mechanism of heterologous resistance of BCG to COVID-19.
About Sandeep Chakraborty:
Sandeep Chakraborty is a researcher & independent consultant who completed his B-Tech in Electrical Engineering from IIT Kharagpur (1997), and Masters in Computer science again from IIT Kharagpur (2008). He has worked in the semiconductor industry for companies like nVidia (Santa Clara, USA) and Mentor Graphics (Noida, India) for more than a decade. Since 2010, he has changed his career course and focused on research in computational protein modelling working with Dr BJ Rao in TIFR, Mumbai. Subsequently, he has collaborated with Dr Abhaya Dandekar (UC Davis, USA) on several projects (Walnut genome assembly, designing new anti-microbial peptides) working at UC Davis as an associate specialist. Currently, he works as an independent consultant.