Objective
Changes in the types of germs (bacteria) in the gut occur before Type 1 diabetes (T1D) development. In people at risk of T1D, we can measure more specific bacteria that contain a product called lipopolysaccharide (LPS) in their cell wall. Importantly the type of LPS made by bacteria that are increased in people at risk of T1D, fails to stimulate protective immune responses, potentially increasing someone’s risk of developing T1D.
We aim to study whether LPS and the proteins that recognise LPS (which hide LPS from the immune system) are associated with both risk of developing T1D and response to a therapy in T1D (a drug called Baricitinib, which can help preserve the insulin-producing beta cells in those with T1D).
We will use a technique called enzyme-linked-immunosorbent assay (ELISA) to measure LPS and a number of different proteins that can recognise LPS from blood samples including TLR4, MD-2, CD14, antibodies and LBP. We will also measure inflammatory and anti-inflammatory immune proteins (called cytokines) to determine whether immune responses to LPS are also altered.
We will determine what factors influence the results we obtain using blood samples from people living with T1D. For example, some samples may have been thawed and frozen repeatedly so we will check that our results are not affected by this process. By understanding these sources of variation, we can make sure that our results are reliable and reproducible before we use very valuable samples.
Next, we will evaluate the role of LPS or the proteins that recognise LPS, in identifying people at risk, and any changes at T1D diagnosis and after this, and will study pre-collected blood samples from 5 groups of children and adults (20-40 individuals in each group):
1. Those at risk of T1D who have not yet been diagnosed with T1D
2. Those at risk of T1D who have since been diagnosed with T1D
3. Those living with T1D for less than 1 year
4. Those living with T1D for over 1 year
5. Individuals not at risk of, or living with, T1D (age- and sex-matched to the 4 groups above)
We hope to study samples collected pre-diagnosis, within 1 year post-diagnosis and then after 1 year post-diagnosis from the same individuals, allowing us to determine how LPS and the proteins we will measure are altered with risk and development of T1D in the same person. This may require us to ask some study participants for a new sample. We will compare the results between the different groups and also identify associations with LPS/proteins that recognise LPS with clinically-available data such as HbA1c, insulin dose, time in/below/above range, age and sex. This may help us to identify whether specific groups of people are more likely to have differences in LPS/proteins that recognize LPS.
We will also determine whether LPS or the proteins that recognize LPS are altered by therapy, specifically a drug called baricitinib, which can impact immune cells, and may alter how immune cells recognize and respond to LPS. We will compare the results between individuals that received Baricitinib to those that did not (placebo). Samples will be studied from individuals before treatment started, and then 3-, 6-, 12- and 24- months after treatment.
Finally, we will use the data we have generated to determine if any of the results we have obtained can identify those who are most likely to develop T1D or who have had a beneficial response to Baricitinib. We hope to be able to use our data to better identify individuals at risk and to ensure people receiving therapy are given the best therapy for them.
Background Rationale
Prior to the development of Type 1 diabetes (T1D), changes in the types of bacteria present in the gut are observed. These changes have been observed not only in individuals at risk of T1D but also those living with T1D. Furthermore, changes in these bacterial abundances appear at a similar time to the development of antibodies that recognise proteins in the insulin-producing beta cells (called autoantibodies). These autoantibodies are currently how we identify individuals who may be at increased risk of developing T1D, and thus, this suggests a link between the immune response predisposing individuals to develop T1D and changes in bacteria being recognised by the immune system.
In humans, increases in the abundance of a specific type of bacteria belonging to Bacteroidetes group are seen in those individuals at risk or living with T1D. These Bacteroidetes bacteria are dominant producers of a product called lipopolysaccharide (LPS), which is found in their cell wall. In addition, a study has shown that the risk of developing T1D was higher in those children whose exposure to LPS was predominantly from Bacteroidetes bacteria, compared to children exposed to LPS from different bacteria called E.coli. The importance of the type of bacteria making LPS was shown in mice, where LPS from Bacteroidetes bacteria could not protect the mice from developing T1D but LPS from E.coli could. This suggests that LPS from Bacteroidetes bacteria, which is increased in those at risk of developing T1D, promotes inflammatory immune responses and fails to induce anti-inflammatory responses (such as IL-10). Thus, LPS from Bacteroidetes bacteria fails to induce protective immune responses, potentially increasing someone’s risk of developing T1D.
Our data have shown that people with T1D have reduced levels of LPS in the blood due to higher levels of proteins that can recognise and bind to LPS, hiding it from the immune system, compared to people without T1D. We hypothesise that this reduced ability of the immune cells to see LPS may reduce the anti-inflammatory immune responses and increase an individual’s risk of developing T1D.
In this proposed study we will investigate LPS and the proteins that bind and hide it from the immune system in the blood of both children and adults at risk of, and living with, T1D and see how these levels compare to age- and sex-matched individuals without risk of, or living with, T1D. Our data will confirm whether we can use LPS or the proteins that hide LPS from the immune system, as a novel way of identifying someone at risk of developing T1D. We hope that this additional way of identifying someone at risk, may help improve those identified to receive therapy to prevent, or delay, the development of T1D earlier.
We will also investigate how LPS and the proteins that recognize LPS are altered in response to therapy, specifically a drug called Baricitinib, which was used in the BANDIT study. This study showed that Baricitinib preserved insulin-producing cells in individuals recently diagnosed with T1D. We chose to investigate our LPS responses in response to this therapy, because when an immune cell recognises LPS, it results in a number of proteins being switched on, including the JAK/STAT proteins; Baricitinib switches these proteins off. Thus, because LPS and Baricitinib work on the same pathway, but in opposite directions in cells, it would be of interest to study whether the changes in LPS or proteins that hide it, are altered in response to Baricitinib therapy. We will use this knowledge to determine whether we can predict if someone is likely to respond well to therapy.
Description of Project
Germs, that include bacteria, live inside our gut, helping us digest food and educating our immune cells to protect us from infection. Prior to the development of Type 1 diabetes (T1D), changes in the types of bacteria present in the gut are observed. These bacterial changes result in people at risk of T1D having an expansion of specific bacteria that contain a product called lipopolysaccharide (LPS) in their cell wall. LPS from bacteria can switch on anti-inflammatory proteins such as IL-10 made by immune cells, which can prevent destruction of the insulin-producing cells. Importantly, the type of bacteria, which are expanded in people at risk of T1D, contain LPS that fails to induce protective immune responses, potentially increasing someone’s risk of developing T1D.
Our data have shown that people with T1D have reduced levels of LPS in the circulation due to higher levels of proteins that can recognize and hide LPS from the immune system, compared to people without T1D. We hypothesize that this reduced ability of the immune cells to see LPS may reduce the anti-inflammatory immune responses and increase an individual’s risk of developing T1D. Our proposed study (over 4 aims) focuses on 2 key questions:
1. Is LPS or the proteins that recognize LPS associated with risk of developing type 1 diabetes?
2. Does measuring LPS or the proteins that recognize LPS help us to identify people responding to therapy (specifically a drug named Baricitinib that alters the immune cells)?
We will do this by investigating LPS and the proteins that recognize LPS from the blood of people at risk of, and living with T1D and compare the results to individuals not at risk or living with T1D. We will also look across time in the same individuals before diagnosis, around diagnosis (within 1 year) and later after diagnosis (over 1 year) to test how these markers change and whether we can see they are linked to T1D development. This study will be completed using mainly stored samples from both children and adults.
We will also investigate changes in LPS and the proteins that recognise LPS in people newly diagnosed with T1D who received a drug called Baricitinib, using pre-collected samples from the BANDIT study. This study showed those taking the drug had better preservation of the insulin-producing cells. We will investigate whether our markers are associated with responses to the therapy as well.
By studying LPS and the proteins that recognize LPS, we hope that we can identify people at risk of developing T1D so that anyone who is at risk can receive therapy as early as possible to try to delay the development of T1D. If we can identify those people earlier, we have a better chance of protecting the insulin-producing beta cells for longer. We also hope that if LPS and the proteins that recognize LPS are associated with responses to therapy, by measuring these we can identify who is responding to treatment. Early identification of those who are not responding can be offered an alternative therapy as early as possible, maximising the ability to protect the insulin-producing beta cells and delaying the development of T1D for as long as possible.
Anticipated Outcome
Our study focuses on measuring a component of the bacterial cell wall (called Lipopolysaccharide; LPS) which is increased in individuals at risk of developing T1D, as well as the proteins that recognise it and hide LPS from the immune system. Studying blood samples from both children and adults at risk of, or living with T1D, as well as age- and sex-matched individuals not at risk, or living with, T1D, we will determine whether:
1. LPS and the proteins that recognise LPS (including proteins called TLR4, MD-2, CD14, LBP and antibodies to LPS) are altered between those with and without T1D, as well as between those at risk of, and living with, T1D
2. Inflammatory and anti-inflammatory Immune proteins (called cytokines), which would be switched on by LPS recognition by immune cells are altered between those with and without T1D, or those at risk of and with T1D
3. LPS, the proteins that recognize LPS and the cytokines switched on by LPS are altered between those receiving Baricitinib (a therapy shown to preserve insulin-producing cells in those with recently diagnosed T1D), compared to those who did not (placebo)
4. LPS, the proteins that recognise LPS and the cytokines switched on by LPS are associated with the presence of other factors such as diabetes risk factors (e.g. genetics, autoantibody presence/levels), duration of diabetes, diabetes management (insulin dose, time in/above/below range).
Together these outputs will allow us to determine whether LPS, the proteins that recognize LPS and the cytokines switched on by LPS can be used as a new way to not only predict those at risk of developing T1D, but also whether someone is responding favorably to therapy. We hope that providing a new way to identify those at risk, we can ensure more individuals a risk can receive therapy earlier, helping to protect their insulin-producing cells for longer and delay the need to inject insulin. Furthermore, we hope that by identifying whether people respond well to therapy, we can ensure individuals receive a therapy that works for them, hopefully delaying the development of T1D.
Relevance to T1D
Our research will benefit people living with Type 1 diabetes (T1D), as well as those at risk, as our proposal focuses on:
1. Developing new ways to both identify those at risk of developing Type 1 diabetes, and
2. Developing new ways to identify whether someone is responding well to therapy
Gut bacteria, which make a component called lipopolysaccharide (LPS) are increased in those at risk of developing T1D and living with T1D compared to those not at risk of, or living with, T1D. Immune responses to LPS have been shown to vary, depending on which bacteria are making the LPS; however, it is clear in those most at risk of developing T1D, the LPS from those individuals fails to prevent the development of T1D. Thus, our study will build on whether this difference in the immune response to LPS can be used to identify those at risk of developing T1D. Our preliminary data, from people living with T1D, show that proteins that bind and hide LPS from being detected by the immune system are increased, compared to those without T1D. Our study will investigate these changes not only in those living with T1D but those at risk of T1D in both children and adult blood samples, as diabetes can develop at any age. We hope to develop a way of improving how we identify someone at risk, by including measurements of LPS, proteins that hide LPS from the immune system, and immune proteins (cytokines) switched on by LPS. By improving our ability to identify someone at risk, we hope that they can receive therapy earlier, delaying the development of T1D for as long as possible.
We will also investigate whether measurements of LPS, proteins that hide LPS from the immune system, and immune proteins (cytokines) switched on by LPS can be used to identify someone who responds well to therapy. We will use blood samples collected from those who did or did not receive a drug called Baricitinib, which can preserve the insulin-producing cells. By identifying ways to study those who respond well, or not well, to therapy, we can ensure those at risk of, or living with T1D, receive a therapy that can protect their insulin-producing cells for longer.
Our ability to predict risk of developing T1D and response to therapy will be determined by the end of this 3-year study. By showing that these markers can be used in this way, we would next wish to understand why people at risk of T1D respond differently to LPS, which may allow us to develop a novel therapy aimed at promoting the correct recognition of LPS to prevent the development of T1D.