TNF, BCG, and You and Me: An In-Depth Look at the Faustman Lab Research


Author’s note: I believe there will be a cure for type 1 diabetes in my lifetime, and I believe it will come from exactly the sort of researcher that Dr. Faustman is—one who drives forward despite doubt from every corner. I also believe that the science and theory behind Faustman’s trials is solid. I don’t, however, think this particular BCG study is well-directed or analyzed, and I think if we want a cure, we have to do better. What follows, then, is not meant to crush anyone’s hopes for a cure, but rather to make sure that we diabetics are closely watching the science that can change our lives. Many of the statements below are my opinion only, and I am only a graduate student. Take it for what it’s worth.


The diabetes world is abuzz with the news that Dr. Denise Faustman has published results from a small trial with promise for a cure for long-term diabetics. Catherine Price did a wonderful job of summarizing the findings and assessing the relevance to diabetics; what follows here is an in-depth look at the paper from the Faustman lab.

TL;DR: There are many studies over many years that support Faustman’s suspicion that inflammation of the innate immune system , the part of the immune system that serves as the body’s first line of defense against pathogens, plays a crucial role in the development and maintenance of autoimmunity. However, the results as reported do not support the claim that the Bacillus Calmette-Guerin (BCG) vaccine had a significant effect on insulin production or T cell viability in subjects.


Part 1: The Background Story

All right, for those of you still with me, let’s begin. In PLoS ONE last week, Dr. Faustman’s lab published results from a human trial that began enrolling patients in 2009; according to the publication, a vaccine used against tuberculosis for almost a century, the Bacillus Calmette-Guerin (BCG) vaccine, showed promise as an immune-modifying treatment for long-term diabetes. Faustman’s use of the BCG vaccine had less to do with the vaccine itself than the fact that the vaccine has been shown to cause cells of the immune system to produce a protein called Tumor Necrosis Factor alpha (TNF-a), which Faustman and other labs have been researching in relation to type 1 diabetes for more than twenty years. So what is TNF-a, and why is it so important?


What is TNF-a?

TNF-a is a protein made by cells of the immune system in response to perceived danger. It is one of the hammers of the immune system; it is a big, blaring alarm signal that tells all surrounding cells that something is wrong and they better start their defense protocols. Not surprisingly, TNF-a is a favorite protein among many scientists– it is important in many types of inflammation, from rheumatoid arthritis to heart disease to cancer [1], and it tends to cause nice, observable responses, which is important when conducting scientific experiments.

However, TNF-a is not a simple on/off switch for inflammation, and it’s important to understand some of the complexity of the TNF-a signaling cascade. To start with, TNF-a is produced by a number of cells, but mostly by macrophages and other cells of the innate immune system. In diabetes, we usually think about T cells, since those are the cells that have been tied to killing beta cells. T cells are part of the adaptive immune system, which helps protect the body by learning to recognize pathogens as they come in, and building new defenses against them. The innate immune system, on the other hand, does minimal learning of new pathogens; instead, cells of the innate immune system recognize certain patterns of proteins that are pathogenic (like flagella from bacteria, for example), and help to activate the adaptive immune system quickly when these patterns are recognized. When cells of the innate immune system see danger signs in the body, they quickly begin to secrete TNF-a and other proteins to signal to the rest of the body that action is required.

These signals get complicated. TNF-a is a big one, but that’s just one of dozens of communicative proteins, called cytokines, that are secreted by macrophages and other innate immune cells when danger is perceived. Any cell in the vicinity of these secreted signals is equipped with many different receptors in differing amounts, all of which pass the signal into the cell and help it “decide” what it should do. And, to make matters worse, it’s not just whether a certain cytokine is there that matters to the surrounding cells; the duration and amplitude of the cytokine signals matter, too. For example, sitting in TNF-a for an hour will produce a different response in a T cell than receiving a TNF-a pulse for ten minutes [2].

To summarize what we’ve got so far: cells of the innate immune system see danger, and secrete a bunch of signaling proteins called cytokines, one of which is TNF-a. And it’s complicated.

This signaling system is important for anything the immune system does, and the development of type 1 diabetes is no exception. Type 1 diabetes is called a T cell mediated disease because T cells are the cells circling around beta cells like hungry wolves as the beta cells die [3], but T cells are not acting alone. Macrophages and other cells of the innate immune system are the first to invade the pancreas in mouse models, and these cells have also been shown to be over-active or functionally deficient in a number of ways [4]. Some researchers theorize that cells of the innate immune system first respond to some natural beta cell death or pathogen in the pancreas, but overreact, leading to a cascade of immune signals that ends with T cells being recruited to kill off all the beta cells [5].

The precise role of each cell type in the pathogenesis of diabetes is still up for debate, but over the years scientists have accumulated evidence that we shouldn’t be myopically focused on T cells. One clear piece of evidence is the role that TNF-a plays in mouse models of diabetes. As elsewhere in the body, macrophages and other cells of the innate immune system are the main producers of TNF-a in the pancreas during the development of diabetes, and there is lots of TNF-a to be found there [6].

And here’s where things get interesting for a cure-seeking diabetic: when scientists saw that cells were flooding islets with TNF-a, they decided to see what would happen if they changed the levels of TNF-a in mouse pancreases– and they found that changing the levels of TNF-a changes whether a mouse will get diabetes.

Now take a guess: we have a blaring distress signal, TNF-a, that turns on all the cells of the adaptive immune system, and we have a disease that is characterized by adaptive immune cells overreacting and killing the body’s own cells. Does adding TNF-a make disease development in mice faster or slower?

As with any good scientific problem, the answer is– “It depends.” In the 1990s, scientists found that if you block TNF-a from birth, diabetes is prevented in Non-obese diabetic (NOD) mice, and if you give the mice recombinant TNF-a at two weeks of age, they get diabetes faster [7]. However, if you give TNF-a to NOD mice at eight weeks of age, disease onset is delayed or prevented altogether [8]! This duality was shown quite elegantly recently, with a strain of mice that could be induced with the drug doxycycline to express TNF-a in beta cells on cue. If TNF-a was expressed early, disease progression was worse; if TNF-a was expressed later, the immune reaction to beta cells was seemingly lessened, resulting in an amelioration of disease symptoms [9].

What’s going on here? What is TNF-a doing that if plays such a bimodal role in the development of diabetes? That’s a good question, and not one that scientists have a clear answer for yet. TNF-a is generally pro-inflammatory, but the signaling systems are full of negative feedback loops to prevent exactly the sort of autoimmunity we see in type 1 diabetes. Perhaps an abundance of TNF-a early in disease progression induces an overreactive T cell response, but later in disease progression leads to an increase in regulation of the immune system, in which the immune system says, “Well, this has gone far enough!” and begins to turn off other cytokines and to kill off activated T cells to prevent overstimulation [9].

Faustman argues that NOD mice have a genetic defect that makes them less able to activate a particular cellular pathway that is important in inflammation and cell survival. Under normal circumstances, TNF-a turns T cells on; however, if there is too much TNF-a, or the cells are defective as Faustman argues the autoreactive NOD T cells are, TNF-a can act as a death signal instead, leading to apoptosis, or self-destruction, in affected cells [10].

Whatever the suggested cause, the take home lesson is that inflammatory signaling is complicated, and the timing and dynamics of signals matter. We can’t, therefore, ignore the many cells of the innate and adaptive immune systems when thinking about diabetes. Further, the more we understand about how these signals work, the more power we have to introduce or remove signals, thereby tipping the balance of the immune system away from autoimmunity.

And what does this all have to do with BCG?

Trying to tip that balance is where Dr. Faustman and the BCG vaccine come in. Faustman’s lab has spent many years trying to understand the causes behind this duality of TNF-a in type 1 diabetes, and has come to espouse the idea that, contrary to expectations, we need to induce the proinflammatory signal TNF-a, not block it, in order to take advantage of the defective pathway Faustman proposed and kill off autoreactive T cells [11]. This is a controversial view, and a high-risk one; treating mice with recombinant TNF-a is successful enough, but treating humans with TNF-a is a whole other story. Given that TNF-a is such an important and pro-inflammatory signal, chronic treatment with TNF-a could be dangerous, and is certainly not FDA approved. Rather than try to get a potentially toxic treatment FDA approved [12], Faustman decided to move upstream, and to treat humans with the FDA approved vaccine for tuberculosis, Bacillus Calmette-Guerin (BCG).

BCG is a vaccine composed of live bacteria, Mycobacterium bovis, that is related to Mycobacterium tuberculosis, which causes tuberculosis. M. bovis is much less potent in humans than M. tuberculosis, and therefore it works well as a vaccine by showing the immune system what M. tuberculosis looks like without presenting any real threat to the human [13].

The way that BCG, and many vaccines for that matter, works is key to Faustman’s logic. When the vaccine is injected into the body, the innate immune cells quickly recognize the foreign bacterial proteins and begin an inflammatory program, secreting lots of cytokines to let the adaptive immune system and other cells of the body know that an invader has been detected. One of the cytokines released is TNF-a [14].

Faustman’s hypothesis, therefore, is that vaccinating long-term type 1 diabetics with BCG will set off a natural release of TNF-a system wide, and this should be a good first approximation of treatment with TNF-a directly [12].

This, however, is where I have my first major disagreement with Faustman’s hypothesis. BCG does cause cells of the innate immune system to produce and release TNF-a; however, that is just one among many immune changes that occur in response to BCG. Many cytokines are released, and it is the mix of these that matters, not the absence or presence of any single protein. Further, the proteins from the injected bacteria circulate and are presented to the adaptive immune system by the innate immune system. Given the complexity and intricacy of the immune system discussed above, it seems a vast simplification to say that BCG treatment is equivalent to TNF-a treatment alone.

Of course, though it is not equivalent, it may indeed serve the same purpose in terms of diabetes. Other cytokines notwithstanding, the full activation of the innate immune system seen with BCG vaccination does seem to temper autoreactivity. More than twenty years ago, studies in NOD mice found that treatment with either Complete Freund’s Adjuvant (CFA), which contains M. tuberculosis, or BCG prevented the development of diabetes in the mice [15]. The cause of this prevention is debated; as with TNF-a treatment, perhaps the activation of the innate immune system initiates negative feedback loops that prevent later autoreactivity [16];  perhaps regulatory or Natural Killer T cells may be involved in dampening the effects of the adaptive immune system [17]; or perhaps the autoreactive T cells are especially susceptible to released TNF-a, as Faustman suggests [10].

The effectiveness of these mycobacterial treatments against diabetes in the mouse has made BCG a very interesting potential treatment for type 1 diabetes. However, it should not be confused with TNF-a treatment alone; if the two were acting through equivalent mechanisms, we would expect to see the same duality we see with TNF-a in the BCG experiments, where early treatment makes the disease worse, but later treatment seems to reduce autoreactivity. Instead, studies with CFA and BCG show that treatment even at three weeks of age can have a preventative effect [15]. What explains this difference? The non-specific immune reaction that is induced with the mycobacterial treatments is in some way skewing the immune system away from autoimmunity in mice, and it is unlikely given the window of effectiveness that we can reduce this skewing to TNF-a alone. TNF-a is likely involved in the response, and there is good reason to think the two means of prevention are working through similar pathways, but we are oversimplifying if we say they are equal.

Well, so what, right? I mean, if it prevents type 1 diabetes, what do I care if we call it a tomato or a tomahto? On the one hand, this is true; if BCG prevents type 1 diabetes consistently in mice, then it is interesting to try in humans, regardless of the cause of the prevention. On the other, if we are mistakenly assigning causality to particular proteins, we run the risk of making the wrong moves experimentally and ending up far away from an effective cure.

We can begin to see the effects of assuming equivalence in the fact that BCG has been investigated in a number of ways already as a potential therapy for type 1 diabetes [18, 19, 20]. What looked promising at first has not held up to repeated testing, leading researchers to conclude that there is a hypothesis there that is important– we can affect the progression of type 1 diabetes by modulating the innate immune system– but BCG isn’t the magic bullet that we’re looking for.

So why does Faustman believe her trial will be different? All the other trials, building off the evidence in animal studies, were focused on early BCG vaccination and recent-onset patients. Faustman, believing that BCG vaccination’s primary role in this case is to induce TNF-a expression, and believing that TNF-a’s primary role in this case is to kill the defective autoreactive T cells after they have developed [12], proposes that treatment with BCG will only work with patients who are already diabetic.

This hypothesis is built on several theories about how BCG is working that, as far as I can tell, don’t hold up against the fact that BCG treatment in animals works early on during disease progression. However, if there’s one thing that’s clear, it’s that the immune system is complicated, so far be it from me to say the Faustman hypothesis is not worth trying. She’s taking a risk, but I commend that; we will need scientists making risky gambles to push the field forward.

Coming next week: Part 2

This article refers to: Faustman DL, Wang L, Okubo Y, Burger D, Ban L, et al. (2012) Proof-of-Concept, Randomized, Controlled Clinical Trial of Bacillus-Calmette-Guerin for Treatment of Long-Term Type 1 Diabetes. PLoS ONE 7(8): e41756. doi:10.1371/journal.pone.0041756

[1] Bradley JR. TNF-mediated inflammatory disease. J Pathol. 2008 Jan;214(2):149-60.

[2] Nelson DE, Ihekwaba AE, Elliott M, Johnson JR, Gibney CA, Foreman BE, Nelson G, See V, Horton CA, Spiller DG, Edwards SW, McDowell HP, Unitt JF, Sullivan E, Grimley R, Benson N, Broomhead D, Kell DB, White MR. Oscillations in NF-kappaB signaling control the dynamics of gene expression. Science. 2004 Oct 22;306(5696):704-8.

[3] Coppieters KT, Dotta F, Amirian N, Campbell PD, Kay TW, Atkinson MA, Roep BO, von Herrath MG. Demonstration of islet-autoreactive CD8 T cells in insulitic lesions from recent onset and long-term type 1 diabetes patients. J Exp Med. 2012 Jan 16;209(1):51-60.

[4] Lehuen A, Diana J, Zaccone P, Cooke A. Immune cell crosstalk in type 1 diabetes. Nat Rev Immunol. 2010 Jul;10(7):501-13.

[5] Schulte BM, Kramer M, Ansems M, Lanke KH, van Doremalen N, Piganelli JD, Bottino R, Trucco M, Galama JM, Adema GJ, van Kuppeveld FJ. Phagocytosis of enterovirus-infected pancreatic beta-cells triggers innate immune responses in human dendritic cells. Diabetes. 2010 May;59(5):1182-91.

[6] Dahlén E, Dawe K, Ohlsson L, Hedlund G. Dendritic cells and macrophages are the first and major producers of TNF-alpha in pancreatic islets in the nonobese diabetic mouse. J Immunol. 1998 Apr 1;160(7):3585-93.

[7] Yang XD, Tisch R, Singer SM, Cao ZA, Liblau RS, Schreiber RD, McDevitt HO. Effect of tumor necrosis factor alpha on insulin-dependent diabetes mellitus in NOD mice. I. The early development of autoimmunity and the diabetogenic process. J Exp Med. 1994 Sep 1;180(3):995-1004.

[8] Jacob CO, Aiso S, Schreiber RD, McDevitt HO. Monoclonal anti-tumor necrosis factor antibody renders non-obese diabetic mice hypersensitive to irradiation and enhances insulitis development. Int Immunol. 1992 May;4(5):611-4.

[9] Christen U, Wolfe T, Möhrle U, Hughes AC, Rodrigo E, Green EA, Flavell RA, von Herrath MG. A dual role for TNF-alpha in type 1 diabetes: islet-specific expression abrogates the ongoing autoimmune process when induced late but not early during pathogenesis. J Immunol. 2001 Jun 15;166(12):7023-32.

[10] Hayashi T, Faustman D. Defective function of the proteasome in autoimmunity: involvement of impaired NF-kappaB activation. Diabetes Technol Ther. 2000 Autumn;2(3):415-28.

[11] Kodama S, Davis M, Faustman DL. The therapeutic potential of tumor necrosis factor for autoimmune disease: a mechanistically based hypothesis. Cell Mol Life Sci. 2005 Aug;62(16):1850-62.

[12] p2 in the paper in question


[14] Mollenkopf HJ, Hahnke K, Kaufmann SH. Transcriptional responses in mouse lungs induced by vaccination with Mycobacterium bovis BCG and infection with Mycobacterium tuberculosis. Microbes Infect. 2006 Jan;8(1):136-44.

[15] Qin HY, Sadelain MW, Hitchon C, Lauzon J, Singh B. Complete Freund’s adjuvant-induced T cells prevent the development and adoptive transfer of diabetes in nonobese diabetic mice. J Immunol. 1993 Mar 1;150(5):2072-80.

[16] Yagi H, Matsumoto M, Kishimoto Y, Makino S, Harada M. Possible mechanism of the preventive effect of BCG against diabetes mellitus in NOD mouse. II. Suppression of pathogenesis by macrophage transfer from BCG-vaccinated mice. Cell Immunol. 1991 Nov;138(1):142-9.

[17] Lee IF, van den Elzen P, Tan R, Priatel JJ. NKT cells are required for complete Freund’s adjuvant-mediated protection from autoimmune diabetes. J Immunol. 2011 Sep 15;187(6):2898-904.

[18] Shehadeh N, Calcinaro F, Bradley BJ, Bruchim I, Vardi P, Lafferty KJ. Effect of adjuvant therapy on development of diabetes in mouse and man. Lancet. 1994 Mar 19;343(8899):706-7. Erratum in: Lancet. 2006 Apr 15;367(9518):1240.

[19] Dahlquist G, Gothefors L. The cumulative incidence of childhood diabetes mellitus in Sweden unaffected by BCG-vaccination. Diabetologia. 1995 Jul;38(7):873-4.

[20] Allen HF, Klingensmith GJ, Jensen P, Simoes E, Hayward A, Chase HP. Effect of Bacillus Calmette-Guerin vaccination on new-onset type 1 diabetes. A randomized clinical study. Diabetes Care. 1999 Oct;22(10):1703-7.

[21] Gosselin J, Menezes J, D’Addario M, Hiscott J, Flamand L, Lamoureux G, Oth D. Inhibition of tumor necrosis factor-alpha transcription by Epstein-Barr virus. Eur J Immunol. 1991 Jan;21(1):203-8.

[22] Gosselin J, Flamand L, D’Addario M, Hiscott J, Menezes J. Infection of peripheral blood mononuclear cells by herpes simplex and Epstein-Barr viruses. Differential induction of interleukin 6 and tumor necrosis factor-alpha. J Clinnvest. 1992 Jun;89(6):1849-56.

[23] Morrison TE, Mauser A, Klingelhutz A, Kenney SC. Epstein-Barr virus immediate-early protein BZLF1 inhibits tumor necrosis factor alpha-induced signaling and apoptosis by downregulating tumor necrosis factor receptor 1. J Virol. 2004 Jan;78(1):544-9.

[24] Haskins K, Cooke A. CD4 T cells and their antigens in the pathogenesis of autoimmune diabetes. Curr Opin Immunol. 2011 Dec;23(6):739-45.

[25] Seyfert-Margolis V, Gisler TD, Asare AL, Wang RS, Dosch HM, Brooks-Worrell B, Eisenbarth GS, Palmer JP, Greenbaum CJ, Gitelman SE, Nepom GT, Bluestone JA, Herold KC. Analysis of T-cell assays to measure autoimmune responses in subjects with type 1 diabetes: results of a blinded controlled study. Diabetes. 2006 Sep;55(9):2588-94.

[26] Kent SC, Chen Y, Bregoli L, Clemmings SM, Kenyon NS, Ricordi C, Hering BJ, Hafler DA. Expanded T cells from pancreatic lymph nodes of type 1 diabetic subjects recognize an insulin epitope. Nature. 2005 May 12;435(7039):224-8.



Notify of
oldest most voted
Inline Feedbacks
View all comments
Bernard Farrell
Bernard Farrell
10 years ago

Thanks Karmel for this detailed explanation. I’m going to read it again because the various biological processes are complicated.@Jo, I spoke with Dr. Faustman about this trial as it was underway. She told me that the FDA allowed it to proceed with a small number of participants because BCG was a long-used chemical, I think they didn’t have to worry as much about possible side effects.I agree with you Karmel, if Dr. Faustman’s work doesn’t result in a cure something along these lines may well do so. Of course the word <strong>cure</strong> is so challenging with everyone taking a different… Read more »

10 years ago

Thanks so much for this – an analysis that doesn’t just look critically at the underlying science, but also makes it comprehensible. The immune system is indeed complicated, and I look forward to the day when we have answers to the questions that you raise.  

10 years ago

I too am a believer in Dr. Faustman’s research.  I too have problems with her most recent published study.  1. there were too few participants to make the study valid.  2.  no study is valid until it can be replicated.  Another study, with many more participants, is needed.  A double-blind study would be prefered.  I look forward to a cure for diabetes in my lifetime, but good research takes time and adequate numbers.

Copyright © 2009-2021 Diabetes Media Foundation, All Rights Reserved.
ASweetLife™ is a trademark of the Diabetes Media Foundation, All Rights Reserved.
Would love your thoughts, please comment.x