Needles be Gone for Type One Diabetes Patients
Final Research Project Update
Update 03/06/2019 (Final Report)
The generous grant from the DRC enabled the critical first step toward a microbial treatment for diabetes that circumvents the struggle of injected insulin therapy. We reached a milestone to demonstrate deep-skin bacteria producing and secreting insulin. Moreover, the knowledge we gained and the materials we generated with the DRC support will strengthen our next efforts, which are now clearly defined because we experienced aspects of the work already. Securing funding for the subsequent work was also an important goal. Regarding this, we are happy to report that our skin bacteria work is part of a project headed by our team member John Glass to be funded by the Larry L. Hillblom Foundation. Therefore, our future efforts will become a reality.
Read the full report here.
Success will bring more NIH funding and could eliminate needles!
Insulin delivery using infusion pumps can be effective for treatment of T1D, but it does not completely protect T1D patients from the long-term effects of the disease or enable a normal non-diabetic lifestyle. Diabetes research has focused on using insulin-producing cells isolated from cadavers or made from stem cells in T1D patients. While some of the research has yielded promising results, nothing yet has radically changed general approaches to treat patients.
Our approach is to give bacterial cells that naturally live in our body the ability to function like our insulin-producing cells, to synthesize insulin when blood glucose levels are high to maintain proper glucose levels in T1D patients. In the proposed project, we will establish the feasibility and safety of this approach by making bacterial strains with this function and testing them in mouse systems. If successful, our research will form the basis for a bacterial treatment that can circumvent the struggle of injected insulin therapy and the issues regarding the rejection of transplanted human cells.
An important ingredient of our study is a set of bacteria recently discovered to naturally reside deep within the human skin, in a layer previously thought to have only human cells. This layer of the skin contains blood vessels and is, therefore, suitable for implanted cells to monitor blood glucose levels and release insulin for systemic distribution. These bacteria do not cause problems in our body. Unlike transplants, they do not trigger host immune response. Unlike insulin pumps, they can enter this deep layer of the skin without puncturing the skin!
Advances in biology now enable making changes to DNA, the genetic material, in these bacteria. Our team at the J. Craig Venter Institute is at the forefront of these advances to produce safe and beneficial bacteria. To make the deep-skin bacteria function like insulin cells, we will introduce a gene into them for making a version of insulin. Bacteria cannot make native insulin by themselves, but there is a type that can be made in bacteria and is as effective as native insulin. We will also introduce a DNA piece containing three genes for making a glucose sensor to control insulin production in the bacteria.
It is critical that insulin-producing bacteria do not infect healthy individuals. Therefore, we will install a mechanism to prevent the bacterial cells from spreading beyond the designated host. We will then paint the mouse skin with our bacteria and see if blood glucose levels drop.
This critical proof-of-concept experiment will tell us if this skin bacteria-based approach has promise and deserves continued support. If we are successful, we will have experimental data that will attract additional funding from the NIH. Establishing a system that can be tested in actual T1D patients will involve many rounds of experimentation and improvement. For example, our future bacterial cells will have more sophisticated safety features. However, tools needed for this approach are already available in basic form. Moreover, the eventual product may be superior to any other products under development. Therefore, we strongly feel that the work needs to be started now toward determining the viability of this approach. Your support could help revolutionize T1D therapy.
Hello, my name is Yo Suzuki. I am an Assistant Professor at the J. Craig Venter Institute. Our institute is known for its expertise in reading and writing genomes, blueprints for life. When writing genomes, our focus is to design and build beneficial microbes. I have been engineering microbes for 11 years, and finally, a connection is made between my skills and the opportunity to contribute to curing type I diabetes. The goal of this project is to create a bacterial strain that can respond to glucose and produce potent insulin analogs in a mouse. Our long-term goal will be to develop an engineered bacterial strain as a surrogate for beta cells in T1D patients. If we are successful, we will have a microbial treatment that circumvents the struggle of injected insulin therapy and problems with a transplant approach. Our project is the first and critical step toward this long-term goal. Our approach is innovative, but many tools to enable this approach are already available, including harmless bacteria that enter the skin in a non-invasive manner and live in a layer of skin appropriate for glucose sensing and insulin administration. Our tools to control these bacteria will only improve. Therefore, we have to start the work now to test the viability of this promising approach. Your support is greatly needed and appreciated.
Update on 4-11-18 (6 Month Progress Report)
The purpose of this project is to test to see if skin bacteria that respond to glucose levels and produce insulin can be engineered and introduced into laboratory animals to reduce their blood glucose levels, as a precursor to establishing engineered skin bacteria as painless substitutes for glucose sensors and insulin pumps. During the first six months, we made rapid progress and established genetic tools for engineering skin bacteria. This accomplishment sets the stage for the second phase of the project where expression modules for insulin genes and regulatory mechanisms for sensing glucose will be developed.
The work performed is summarized under the aims proposed for the project.
Aim 1. Establish genetic engineering tools in selected deep skin bacteria.
We collected six Gram-positive and three Gram-negative strains of bacteria isolated from skin samples. We determined the minimal inhibitory concentrations for nine antibiotics commonly used for genetic engineering. This test revealed that there are many strain-antibiotic combinations that can be used in our study. The Gram-positive strain Staphylococcus epidermidis ATCC12228 was the only strain tested among the nine strains for the capacity to re-enter the skin (unpublished result, R. Gallo). Therefore, we focused our resources on this strain. Gram-negative bacteria have been noted to be more proficient at proper folding of secreted proteins derived from other organisms, although there are examples where Gram-positive cells were used for making and secreting insulin. We decided to keep the Gram-negative strains as backup strains and continue to acquire genetic engineering tools for them. This marks the attainment of an initially proposed milestone (selection of strains).
We previously developed a genetic engineering approach where synthetic DNA fragments loaded with an enzyme called transposase in vitro is introduced into bacteria. Transposase facilitates the integration of DNA fragments into the genome. Because this approach is effective in a wide variety of organisms, we tested it in the S. epidermidis strain with limited tools. To cost-effectively perform this procedure, we purified transposase and confirmed the activity of the purified enzyme. We then used the enzyme and succeeded in introducing a DNA fragment only containing a puromycin resistance gene into the S. epidermidis strain. The positive result was obtained when the DNA sample was pretreated with the lysate of the organism. The idea was that methylases in the lysate generated the methylation pattern on DNA found in the native organism so that the incoming DNA was accepted as its own DNA.
Aim 2. Implement a biocontainment measure.
Before we engineer the capability to express and secrete insulin in the S. epidermidis strain, a mechanism for biocontainment needs to be implemented, as an organism capable of expressing human insulin would be a hazard to laboratory workers, should they become infected. We are attempting to use CRISPR genome editing to knock out the thyA gene needed to make thymidine, an ingredient for DNA, to make our bacterial cells dependent on thymidine supplied from outside and keep the cells within designated culture tubes.
Aim 3. Express SCIs in deep skin bacteria.
Recent studies have resulted in single-chain insulin analogs (SCIs) that match native insulin in potency (Hua et al., 2008, J. Biol. Chem. 283:14703-14716). In beta cells, proinsulin is folded and cleaved to produce biologically active insulin, which consists of two peptides that are linked by disulfide bonds. Because native bacteria cannot cleave proinsulin, SCIs are essential for the strategy of cell-intrinsic and self-sufficient production of active insulin within bacteria. We are currently designing expression constructs incorporating the published SCI-57 design (Hua et al., 2008).
Aim 4. Evaluate the capacity of bacterially produced SCIs to stimulate glucose uptake in adipocytes.
We plan to test whether the bacterially produced SCIs are biologically active by treating mouse adipocytes with the bacteria in a glucose uptake assay. We expect this work to be started as soon as the SCI-producing strains are made. We have culture cell expertise needed to prepare the adipocytes.
Aim 5. Evaluate the ability of SCIs to reduce blood glucose in mice after application of the engineered bacteria to the mouse skin.
We plan to determine if the SCI-producing bacteria can colonize the skin of a mouse to result in a reduction of blood glucose levels in the mouse. We will prepare for this work when the SCI- producing bacteria are made.
Aim 6. Implement a glucose-mediated regulation of SCI production.
For effective blood glucose control, it is critical that SCI production be adjusted based on glucose concentration. We initially proposed to use a glucose sensor in Gram-negative bacteria, but we decided to focus on a Gram-positive strain. We will start researching mechanisms available in Gram-positive bacteria to enable a smooth transition to this phase of research.
The Diabetes Research Connection grant enables the critical first step toward a microbial treatment for diabetes that circumvents the struggle of injected insulin therapy. The preliminary data obtained in this project were incorporated into a grant preproposal submitted by J. Glass (JCVI) to the Larry L. Hillblom Foundation. We have identified a suitable NIH grant mechanism (PAR-18-434) to further develop our approach. We also attended the 2017 Chemical and Biological Defense Science and Technology Conference in Long Beach, California sponsored by Defense Threat Reduction Agency. Skin biology is an important aspect of their portfolio, and we received useful feedback on our research from experts.
Update on 2-06-18
Last time I told you that I purified an enzyme called transposase to be used for facilitating a process called transformation, to put engineered DNA constructs into cells from outside. It turned out that the transformation step was still not easy for skin bacteria, but by communicating with scientists from Australia and multiple universities in the US working on related bacteria of the Staphylococcus species, I got better at the process. I started getting colonies, or dots in a Petri dish each originating from a single transformed cell. Tests confirmed that these colonies had the DNA material I introduced into the cells. With this process being established, the next step for me will be to knock out a gene (thyA gene) needed to make an ingredient for DNA, to make our bacterial cells dependent on the ingredient (thymidine) supplied from outside, so that we can keep the cells at designated sites like a culture flask (as opposed to my skin). Also, the time is right for making a DNA construct for expressing insulin to be introduced into the skin bacteria. Designing DNA constructs is what I love the most. I am grateful that your support enables us to proceed to this phase of the project.
Update on 11-22-17
Toward my aim 1 goal of establishing genetic engineering tools for skin bacteria, I wanted to test the approach of introducing synthetic DNA fragments loaded with transposase in vitro. Transposase facilitates the integration of DNA fragments into the genome. I believe that this is a widely applicable approach that is suitable for our project with multiple target organisms to engineer. For optimizing this process, I would need quite a bit of transposase, but this is an expensive reagent if you buy it. Just 10 µl of it costs $500, and you can use up all 10 µl in one experiment. Therefore, I decided to make my own preparation of transposase. I received a DNA construct containing a transposase gene that can be expressed in E. coli from a research group in Sweden. I introduced this construct into an E. coli strain suitable for recombinant protein production. I induced transposon production, lysed the cells, and purified the protein from the lysate using an affinity chromatography column. I confirmed the activity of the purified transposase using a standard assay in our laboratory. I succeeded in this process and obtained an amount of transposase worth $93,000. Encouraged by this result, I am now tackling establishing transformation protocols for skin bacteria.