Clinical studies have been initiated worldwide to determine whether BCG vaccination contributes to the reduction of COVID-19 severity. In a previous blog, I hypothesized that BCG inoculation, mainly for classical acquired immunity, would act to suppress COVID-19. In this article, I would like to consider the immune mechanism from the aspect of boosting innate immunity memory based on relatively new literature.
Incidentally, this article is not a recommendation to try BCG in the general medical community for prevention of new coronas for purposes other than conventional tuberculosis prevention.
What is epigenetic?
The word "genetics" can be translated as “遺伝学”, but when you add "epi" to it, it means "after ~". In other words, if you force the word "epigenetic" into Japanese, it becomes “後成的遺伝上の”. The term epigenetic has been around for a long time, but the idea in the field of research has been generally active since around 2000.
Vertebrates and microorganisms, including humans, have cells with nuclei and DNA (deoxyribonucleic acid) stored in them. DNA molecules are usually distributed in the nucleus as fibrous structures wrapped around proteins called histones, and the genetic information is contained in DNA as a polymer.
Source: National Cancer Center Japan
This DNA is something that every living organism is born with, and its genetic information is unchangeable if nothing else. However, when a certain part of DNA is methylated and marked/labeled, or a chemical decoration occurs in the surrounding protein, molecules and proteins corresponding to the gene are generated acquired. This is called an epigenetic change or gene expression. A chemically decorated genome is called an "epigenome". This gene expression is either up-regulated or down-regulated by acetylation or deacetylation of histones. Epigenetic reprogramming means that these DNA markers are erased and reconstituted. Epigenetic reprogramming is said to be involved not only in DNA methylation, but also in the regulation of RNA expression (microRNAs and long non-coding RNAs).
Dynamic innate immune memory
In the classical interpretation of immunology, immunity was divided into "innate immunity" and "acquired immunity", and the division of functions was roughly considered as follows.
In "innate immunity," macrophages (Mφ), dendritic cells (DC), mast cells, and natural killer cells (NK cells) non-specifically phagocytose pathogens and, in some cases, present antigens that are characteristic of the pathogen.
In "acquired immunity", T cells and B cells receive signals from the innate immune system and recognize the characteristics of invading pathogens to activate killer cells and mass produce antibodies to repel them. Some cells remain in memory of the pathogen's characteristics and prepare for the next response. (this is the vaccine's effect.)
Considering the above roles, it can be interpreted that it is in acquired immunity that pathogen characteristics are remembered, but recent studies have shown that memory is not only present in acquired immunity, but also in innate immunity. While the classical interpretation is that innate immunity is considered a relatively static response, in fact, it is increasingly thought to be a dynamic and interactive response occurring, such as remembering pathogen features or boosting innate immunity in a non-specific way to pathogens. Epigenetic reprogramming is said to be involved in the mechanism that causes such a response.
It is increasingly believed that BCG inoculation not only acquires immunity against Mycobacterium tuberculosis, but also strengthens a dynamic innate immune system that responds to more than just bacteria.
Let's organize a little prerequisite knowledge to explain this. Actors (mature cells) such as Mφ, DC, T cells, and B cells, which appear in immunity, are included in white blood cells and living tissues, but they did not originally exist independently. These cells are differentiated from their parent's cells, called progenitor cells.
For example, if we classify progenitor cells by Mφ and trace them back to their ancestors, we get the following HSC (HSC)←Pluripotent progenitor cell (MPP)←granulocyte-macrophage progenitor cell (GMP)←granulocyte-macrophage-forming cell (CFU-GM)←macrophage colony-forming cell (CFU-M)←monocyte←Mφ
(Check out this site and the literature if you want to know more.)
The cytokines produced by each differentiation are various and complex. Putting aside the details, the point is that there are HSCs and MPPs in the bone marrow, and in the case of Mφ, these are the primary mature cells.
Now back to BCG. This paper shows that BCG (BCG Tice strain) stimulates HSCs and MPPs to produce epigenetically remodified Mφs in a mouse model (*1).
(Left) Differentiation of hematopoietic stem cells into monocytes (Right) Changes in the immune response due to external stimuli such as vaccines
The proliferation of leukocyte cells produced from HSCs and their differentiated cells by vaccination involves stimulating factors such as pattern recognition receptors (PRRs), cytokines such as IFNγ, and G-CSF.
Since HSCs are self-renewing cells, if vaccines such as BCG indirectly leave some mark (epigenomic labeling) on HSCs that characterizes the differentiated cells, it is not surprising that BCG may have an innate immune memory against common pathogens. This paper shows that the up-regulated genes in both HSC and MPP were expressed in the BCG-inoculated mice compared to the non-BCG-inoculated mice. Interestingly, on the other hand, the genes involved in differentiation into lymphocytes were down-regulated. Mφs differentiated from HSCs are more protective in mice inoculated with BCG (using mice depleted of T cells in the bone marrow, just in case) against intentional infection by Mycobacterium tuberculosis. And such Mφs have been shown to have high gene expression levels that produce the pro-inflammatory cytokines IFNγ, Tnf, and IL-1β, and the expressed genes are H3K4me3 and H3K27ac. →(※A)
In another literature, these H3K4me3 and H3K27ac are also involved in the production of DCs from monocytes, independent of BCG. (*2)
This suggests that after the activation of innate immunity, DCs may be activated at the same time when Mφ is activated using these expressed genes as a key, and then quickly transition to acquired immunity through antigen presentation. →(※B)
H3K4me3 and H3K27ac represent methylation of histones and acetylation of histones, and H3K27ac represents acetylation of histones. This seems to indicate a state of being ready to have cell differentiation quickly activated/inhibited. (*3)
The details are beyond my comprehension, so I won't go into any more depth here. (Please let me know if anyone knows more.)
The above is not all the explanation for BCG enhancing innate immune cells such as Mφ, but it seems to explain one mechanism that cannot be derived from the classical interpretation.
The effect of BCG vaccination on the immune response
Mechanism of BCG inoculation in the Mφ
Since BCG is a bacterium that has weakened the tuberculosis bacteria, the immune response of BCG in the body should be similar to the internal mechanism of the tuberculosis bacteria when they are infected. Although the bacterium tuberculosis has been known for more than 100 years, its amazingly complex mechanisms in the body have only recently become known. For details, please refer to the literature here (*4), but here is a part of it. When common bacteria enter the body, they are eaten by phagocytic cells such as Mφ, which has a digestive organ (phagosome) similar to the stomach and is broken down when pathogens enter this digestive organ. However, the tuberculosis bacteria have the ability to multiply for a certain period of time in Mφ to prevent it from being digested by any means after being eaten by Mφ. Pathogens that enter the phagosome are usually degraded by enzymes contained in the lysosome to form the phagolysosome in the Mφ. In the case of Mycobacterium tuberculosis, however, LAM and metalloproteinase (Zmp1) produced by Mycobacterium tuberculosis inhibit the activation of caspase-1 in the Mφ, thereby inhibiting phagolysosomal fusion. Caspase-1 has the effect of processing the IL-1β precursor pro-IL-1β and making it ready to secrete IL-1β. (*5)
In this way, tuberculosis bacteria can survive in Mφ, and subsequently infected Mφ activates Mφ by producing cytokines such as TNF-α, IL-1, IL-12, and IL-18 by various mechanisms.
[added on May 27, 2020]
The uptake into the phagosome described above occurs by a mechanism called autophagy induction (*7). After the formation of autophagosomes and autolysosomes, the tuberculosis bacteria are degraded, but it is not known if they are antigen-presented after degradation. If antigen presentation occurs after tuberculosis bacillus degradation by Mφ or DC, there may be a common antigen after COVID-19 is degraded by a mechanism similar to that of the antigen. If this is correct, it would be conclusive evidence that BCG has an effect on COVID-19.
Influence on T-cell-associated infection defense responses
In experiments using mice, it has been shown that when mice are fed BCG, they produce cytokines and chemokines such as IFNγ and TFNα and activate inflammatory cells such as Mφ and neutrophils, and express PD-1L, a substance that binds specifically to PD-1. PD-1 is a receptor expressed on activated T cells and its gene, which was discovered in 1992 in Honjo's laboratory at Kyoto University. This paper shows that when PD-1L binds to the PD-1 receptor, the inhibitory signal inhibits Th1 type T cell function and suppresses the excessive inflammatory response. PD-1 is a substance that has been implicated in cancer immunotherapy, but we have not seen many papers on other types of pneumonia, so we may learn something by confirming the extent to which BCG vaccination status affects the expression of PD-1-related signals after a long period of time.
IL-1β in innate immunity
IL-1β is a cytokine often found in papers describing innate immunity, and in the context of BCG, this paper also describes IL-1β as playing an important role in training immunity. In fact, when monocytes in the test tube were pre-stimulated with IL-1β, they expressed the genes H3K4me3 and H3K27ac, which subsequently protected them from external viral infection. (*6)
Here again, a strong relationship between IL-1β and H3K4me3 and H3K27ac is evident.
Differences in innate immune enhancement by sub-strains
Here again, let's review the experimental results of the paper I mentioned in my previous blog. According to the paper, the cytokines produced when inoculated with the BCG Tokyo strain were IL-1α, IL-1β, IL-6, and IL-24.
If the BCG Tokyo strain has the most enhanced innate immunity of the sub strains, the most likely scenario would be that IL-1β would be the key. In other words, after the BCG Tokyo strain inoculation, IL-1β is produced in large numbers along with the mechanism of infection with Mφ, and gene expression occurs, resulting in a particularly enhanced innate immunity. In this case, not much antibody is produced. When humans are infected with the virus, Mφ is activated by the mechanism in (※A) to phagocytose and repel the virus. Even if it takes a long time to fight off the virus, the person does not become seriously ill because acquired immunity centered on T cells is quickly activated by the mechanism of (*B), and antibodies are produced to drive out the virus.
In other strains than BCG Tokyo, IL-1β is not so strongly activated but IFNγ is produced, acquired immunity may respond normally and resolve with moderate symptoms. When BCG is inoculated, it invades Mφ through the receptor called NOD2 in Mφ, but the amount of IL-1β secretion may be different due to the difference in toxicity between the sub-strains.
Of course, the mechanism of innate immunity (※A) does not describe anything about infection other than that of Mycobacterium tuberculosis, so I don't know if it applies to COVID-19. A similar verification can be confirmed in experiments with SARS-CoV-2 and mice. In addition to BCG Tokyo, it may be possible to express IL-1β for a longer period of time in sub-strains other than BCG Tokyo with the same level of virus protection. (I'd love it if someone could verify it.)
The above is a recent immunological perspective on the protection against COVID-19 by BCG vaccination. There is not much evidence on the memory of innate immunity, and I feel that it is still necessary to investigate the mechanisms of gene expression and its long-term maintenance during differentiation from HSC to mature cells. Considering this, we may be in a time when a paradigm shift from the classical interpretation of acquired immunity to an additional perspective of innate immunity may be needed.
Data support the hypothesis that BCG vaccination reduces the severity of the COVID-19 (hereafter referred to as the BCG hypothesis). The causal relationship between BCG and the COVID-19 is expected to become clearer in the future, and I would like to investigate the possibility of this causal relationship in my own way, formulate a hypothesis, and organize it.
I am not an expert in medicine, biology, or immunology, so the following description may contain parts that are professionally inaccurate. If you want to know more, please check the information sources described by the experts. This article has not been reviewed by an expert at this time. Any comments from experts would be appreciated. This article does not recommend that BCG be attempted in the general hospital for the prevention of COVID-19 infection for purposes other than conventional tuberculosis prevention.
What has been revealed about the death toll of the COVID-19
Mortality in statistics
There is a significant difference in mortality per unit of population in countries where BCG vaccination is mandatory, where it was mandatory in the past, where it is no longer mandatory, and where it is no longer mandatory.
There is a significant difference in mortality by BCG strain type: BCG Tokyo has the lowest mortality rate, followed by other BCG strains; BCG-naïve countries have higher mortality rates. (*1-1, 1-2)
Trends in COVID-19 infections
It has been observed that the following infectious trends of the COVID-19 are different from those of normal infections.
The COVID-19 includes a certain percentage of people who are infected but don't show any symptoms.
There may be respiratory abnormalities without a fever.
Even when infected, the rise in antibodies may be gradual.
Even if you get infected and recover, your antibodies may fall off.
What I want to investigate
I investigated the hypothesis that BCG is effective against the COVID-19 and why older strains such as BCG Tokyo are more effective against the COVID-19 than other strains, mainly from the viewpoint of immunology, as follows.
Immunity and BCG
Immunity is the ability to keep one's body normal by attacking bacteria and viruses that have invaded the body as abnormal substances.
The human immune system is divided into "innate immunity" and "acquired/adaptive immunity". Innate immunity refers to immunity that a person is born with, while acquired immunity refers to immunity acquired after birth through infection or vaccination. The general flow of a person's immune response is as follows.
(1) Pathogens enter the body and innate immunity is activated (macrophages, dendritic cells and mast cells fight against pathogens).
(2) T cells proliferate after stimulation by foreign proteins (collectively called antigens) of foreign pathogens, and after eliminating pathogens, some of them differentiate into antigen-specific memory T cells. These memory cells remain, which allows for a rapid immune response when re-exposed to pathogens.
When we recognize a pathogen in such an immune response, we release substances called cytokines to call on cells to help us attack the pathogen. There are various types of cytokines such as interleukin (IL-n) and IFNγ, and in the case of innate immunity, cytokines such as TNF-α and IL-1 are produced.
It is said that BCG vaccine stimulates natural immunity, which is a state of being trained to make it easier for natural immunity to work, and this may take root as a long-term memory. This is called training immunity.(*2)
Mechanisms of viral infection and onset
The following is a brief description of how infection with a new type of coronavirus, an RNA virus, can lead to serious illness.
The steps of viral infection and propagation
(1) More than a certain number of COVID-19 viruses enter the body through droplets or contact (*3)
(2) The virus that enters the body adsorbs the receptor called ACE2 in the body's cells.
(3) The viral envelope fuses with the cell, and the viral protein and nucleic acid (RNA) enter the cell.
(4) Breaks down the protein that encases the viral RNA and releases the RNA.
(5) Replicate the released RNA.
(6) Viruses replicate with replicated RNA and proteins.
Source: The Institute of Medical Science, The University of Tokyo
Viral uptake and infection occur mainly in the upper or lower respiratory tract, and proliferation occurs in the lungs from the lower respiratory tract onward and in organs with ACE2 receptors. During the process of proliferation, natural immunity is triggered and cells such as macrophages in the body attempt to phagocytose the virus. At the same time, macrophages release substances called cytokines to get support from other cells, activating immune cells (such as killer cells and antibodies). Normally, phagocytosis by innate immunity leads to a decrease in viral cells and healing, but if this cannot be suppressed, cytokines are abnormally activated by other substances, causing a state of inflammation (this is called a cytokine storm). A cytokine storm is a condition in which "release of cytokines" → "activation of immune cells" → "release of cytokines" are repeated and become abnormally active. It eventually leads to respiratory distress and acute respiratory distress syndrome (ARDS).
BCG-induced immune response
There are various pathogens and viruses in the world and they are taken into the body unknowingly, but we usually stay healthy because the immune cells in the blood attack the pathogens and get rid of them. The main types of immune cells are macrophages and dendritic cells differentiated from monocytes, which are a type of white blood cell.
Well, as I've mentioned a bit on the blog before, an interesting paper was published in 2018 by a Dutch research team related to training immunity with BCG vaccination (*4). Inoculation with BCG induces a switch in the immune cells as a gene ornament to a state in which they can easily activate (this is described as induction of epigenetic reprogramming), and cytokines (IL-1β) in immune cells rise against the pathogen yellow fever virus (YFV), and when compared with inoculation without BCG, a decrease in the virus in the blood was observed five days after inoculation with BCG, confirming the phenomenon of protection from infection.
Source: Cell Host & Microbe
In this experiment, the yellow fever neutralizing antibody titer after 90 days of vaccination was not significantly different from that of BCG-free vaccination.
The "state in which immune cells tend to activate" is specifically a state in which histone proteins contained in monocyte chromatin (*5) activate genes, which can be determined by analyzing the genome sequence before and after BCG inoculation. The paper also argues that abnormal production of IL-1β may play a central role in inducing training immunity.
This paper does not validate against the COVID-19, but it does provide an example of how the BCG vaccine may be effective against a specific virus, and if we can see the similarity between FYV and the SARS-CoV2 virus, we may be able to say more with certainty.
How to recognize pathogens
Recognition by Pattern Recognition Receptors
The human body has the ability to recognize various bacteria and viruses that are lurking in the body and whether they are foreign or not. There are several types of PRRs, including Toll Like Receptors (TLRs) on the cell surface and intracellularly, and RIG-I-like Receptors (RLRs) intracellularly, which are known to accept viruses.
Source: Jpn. J. Clin. Immunol., 34 (5) 329~345 (2011) (C) 2011 The Japan Society for Clinical Immunology
TLRs are classified into 11 types, among which TLR-7 and TLR-8 recognize single-stranded RNA viruses (*6, 7). These TLRs respond to the RNA or DNA of the virus rather than the virus itself.
Dr. Hirano at Osaka University has reported that RIG-I and MDA5 are the mechanisms that recognize SARS-CoV antigens in innate immunity. It is also explained that ARDS is caused by a cytokine storm centered on IL-6. (*8)
Differences in characteristics between BCG strains
(1) Differences in ingredients between sub-strains
BCG was introduced from the Pasteur Institute in 1926, and subspecies are divided by country, with strain-specific gene deletions and mutations occurring during each country's own culture. The BCG Tokyo strain in Japan is based on the original bovine bile (M.bovis) and was distributed relatively old. Each subspecies has been found to have different bacteriological traits as shown in Table 1.
Source: History of BCG. How and what should we learn from its history.
MPB and IS6110 and other gene regions called RDx (Region of Difference) remain in relatively old Tokyo/Moreau/Russia strains, while MPB and IS6110 and other gene regions called RDx remain in relatively old Tokyo/Moreau/Russia strains.
(2) Differences in epitopes
Another way to recognize pathogens is through epitopes. An epitope is the part of an antigen (protein) that an antibody recognizes, and usually one antigen has multiple epitopes. When an antigen is recognized by a T cell, the antigen is first incorporated into macrophages and antigen-presenting cells of B cells, where it is degraded into peptides. The processed peptides are presented to the T cell receptors and the T cells are activated. (*9)
In 2013, a Chinese research team looked at how BCG's genomic information and epitope counts differed from strain to strain of BCG. Out of a total of 483 T cell epitopes, older generation lines such as BCG Tokyo/Russia have been shown to have a higher number of epitopes. (*10)
Fig 1. Different number of epitopes
Does this mean that the number of epitopes contained in MPB is included in this group? The paper also suggests that the BCG Tokyo strain is the best candidate for use in better vaccine development.
Differences in immune response between sub-strains
How does the immune response change after inoculation with BCG, which has different characteristics depending on the strain as described above? Here are some papers that may be helpful. (*11)
This paper examines the immune response from peripheral blood mononuclear cells (PBMCs) collected one year after neonates were inoculated with BCG Brazil, BCG Denmark, and BCG Tokyo.
When the cytokines expressed were divided into three types, the BCG strains showed the following results.
Type
Cytokine
BCG strain
Type I
IL-1α, IL-1β, IL-6, IL-24
BCG Tokyo
Type II
IL-2β(IL-12β), IL-27, IFN-γ
BCG Brazil or BCG Denmark
Type III
TypeI cytokine+TypeII cytokine
BCG Denmark
It is natural that BCG vaccination activates acquired immunity and produces various cytokines. From the above results, the following points are noteworthy.
The BCG Tokyo strain expresses IL-6, which is found in cytokine storms seen in cases of the COVID-19. (But IL-6 is not unique to the COVID-19.)
The BCG Tokyo strain expresses IL-1, which is produced by innate immunity. It's possible that he's boosting his natural immunity.
Memory cells in immunity
Memory T cells, memory B cells, and memory NK cells are known to be involved in the mechanism of recognizing and storing antigens.
Once a human being is directly exposed to an antigen, it recognizes and remembers the antigen, so that it can quickly prepare for an attack when the antigen comes again. It has been suggested that the vulnerability of young children to infection may be due to the lack of formation of this immune memory. In other words, "we have a lot of naïve cells that have no memory.
Source: Cancer immunity.jp (ONO PHARMACEUTICAL CO., LTD)
Diverse defense mechanisms against pathogens
In addition, CD4+ memory T cells are responsible for memory against antigens derived from the virus. Interestingly, when memory T cells respond to the influenza vaccine, they also respond to environmental antigens other than influenza, suggesting that CD4+ memory T cells have a broad antigen response (cross-reactivity is said to be high) and that vaccinees have many such memory T cells. (*12)
Naïve T cells take a few days to a week or more to respond to an antigen, but when memory cells are formed, an antigen response can occur in a few hours.
There are two types of memory B cells, high affinity and low affinity, which are more likely to respond to similar antigens than to specific antigens. These memory B cells are produced by helper T cells (*13). Memory B cells have been shown to be activated via IL-9, which may not be relevant because BCG inoculation does not produce IL-9.
Incidentally, studies have shown that BCG inoculation in adults sustains NK cell responses for a long period of time. (*14)
Possible hypothesis
Based on the above paper, a hypothesis will be developed to test the effect of BCG vaccine in reducing the severity of COVID-19.
People who have not received BCG first activate innate immunity, but either they cannot present antigens and do not produce antibodies, or their lymphocytes do not become activated after presentation of antigens and antibody production is delayed for more than a few days, resulting in rapid viral proliferation and a strong cytokine storm, which leads to severe disease.
After activation of innate immunity, memory T cells rapidly recognize antigens and activate lymphocytes to produce antibodies, which suppresses viral proliferation to a certain degree, and cytokine storms are moderately suppressed in BCG-treated patients, making them less susceptible to severe disease.
In addition to the above, people who have been inoculated with old strains such as BCG Tokyo rapidly produce IL-6, and innate and acquired immunity are activated almost simultaneously, stopping viral proliferation quickly, and cytokine storms are unlikely to occur. Some people will be cured without another rise in antibodies.
Although the above is the overall hypothesis, the following immunological events can be considered.
Hypothesis: What might happen in the case of BCG vaccination?
A significant difference in the increase in cytokines such as IL-1β in blood monocytes of BCG-inoculated individuals was observed when the new coronavirus was administered to blood samples from BCG-inoculated or non-BCG-inoculated individuals.
At the same time, cytokines in these monocytes are caused by the activation of lymphocytes such as memory T cells (CD4T system, CD8T system).
Hypothesis: What would happen if there is a difference in efficacy between different BCG sub-strains?
Epitope analysis of BCG revealed similarity to the antigens of the new corona (which could be said to be prone to cross-reactions). In particular, many similar epitopes were found in BCG Tokyo strains especially MPB64, MPB70 and MPB80. (Would such similarities be known by analyzing the peptides in the epitope mapping?)
There is a certain correlation between the index of similarity and the number of epitopes in Figure 1.
In the BCG Tokyo strain, IL-6 expression activates Tfh (differentiated T cells that target B cells) and B cells at an early stage.
Although not all of the BCG hypotheses can be proven, I believe that experimentally confirming the above hypotheses will help to reinforce the BCG hypothesis. In particular, the long-term memory of the memory cells of lymphocytes and the memory mechanisms of innate immunity have been partially discovered, but the details are still not clear. It is hoped that these basic studies will lead to a better understanding of this phenomenon.
