Designing Enzymes to Increase Red Blood Cell Shelf Life

More than 4.5 million patients need blood transfusions each year in the US. In the US, 114,000 pints of blood expire annually and are unable to reach patients.

The reason is, that hemoglobin in blood donations oxidizes at a faster rate in storage compared to in vivo, this causes oxidative stress to break down red blood cells at a faster rate, leading to a quicker expiration date of blood.

We are in a blood shortage crisis

Every two seconds in the United States, someone will need a blood transfusion. This makes up approximately 1 of 7 people entering a hospital, approximately 5 million people.

One pint of blood can save up to 3 lives. However, in the United States, 114 thousand pints go unused due to a fast-approaching expiration date. Even so, millions still don’t receive blood transfusions, but why?

The Problem with Blood Storage

Worldwide, red blood cells are the most widely transfused blood component.

Currently, the most widely used protocol for the storage of red blood cells (for up to 42 days) is the collection of blood into anticoagulant solutions, usually citrate-dextrose-phosphate.

Red cell concentrates are prepared by the removal of plasma and sometimes also leukoreduction. The product is stored at 2–6° C in a slightly hypertonic additive solution.

The FDA has indicated that red blood cells can be refrigerated for a maximum of 42 days. However, a study published in Anesthesia and Analgesia indicated that after 21 days, the membranes of stored red blood cells have stiffened — a result of their aging and interaction with a new medium making them unusable.

How Blood Cells Blood Decay

Red blood cells undergo biological changes as they age, including the shedding of the nucleus and ribosomes, and the aging of their hemoglobin.

Hemoglobins are the hundreds of millions of proteins within the blood cells that carry oxygen within the red blood cells throughout the bloodstream. As hemoglobin bonds with oxygen over time, the polar-covalent force of the Iron-Oxygen bond pulls the Iron’s electrons into Iron’s third oxidation state (Iron III), leaving the hemoglobin unable to bond with oxygen.

The result of the Iron III trying to react with oxygen produces oxygen anions (O-) that react with surrounding particles to form toxins like hydrogen peroxide (H2O2) that lead to the natural decay of the red blood cell’s cytoplasm and membrane.

The body has a natural way of dealing with the natural oxidation of hemoglobin, using flavoenzymes to reduce the oxidized Iron lll back to its natural Iron ll oxidation state, thereby reversing the aging of the hemoglobin.

Peroxidase enzymes are also used to break down the toxins that are produced by the oxidized hemoglobin.

How the Rate of Decay Increases Outside of the Body

The refrigerated temperature at which donated blood is stored (~4deg C) to prevent bacterial infections significantly slows the metabolism of the red blood cells.

The slowed metabolism significantly lowers the rate at which the enzymes can reverse the effects of the hemoglobin aging, but as the hemoglobin still bonds with oxygen, the rate at which the hemoglobin oxidizes becomes faster than the rate at which the enzymes can keep up.

The deficit of oxidized-hemoglobin-reducing enzymes allows the processes of oxidative damage to grow fast enough to significantly shorten the lifespan of the red blood cells.

How People are Preventing this Discrepancy

Artificially preventing the enzyme deficit by increasing the count of enzymes per cell, after a donation, would allow for the oxidized hemoglobin to be reduced at the same rate as within the human body, or even faster.

Reducing the oxidized hemoglobin will slow the creation of the oxygen-reactive species toxins (like H2O2) that lead to the decay of the red blood cell’s cytoplasm and membrane.

Furthermore, the introduction of more enzymes that break down the toxins, like peroxidase enzymes, can decrease the number of toxins that will damage and decay the cells.

In total, the approach of speeding up the slowed enzyme processes in refrigeration that are responsible for enzyme upkeep can hypothetically increase the lifespan of donated blood significantly beyond 42 days.

Professor Joshua Welch — Assistant Professor of Computational Medicine and Bioinformatics

Currently working on the application of HBOCs in blood and advocates that they are looking into them due to promising applications of preventing the degradation of hemoglobin.

Current Gaps in the Status Quo

1. To solve the issue research companies have been looking more into HBOCs (hemoglobin-based oxygen carriers) which can act as peroxidases which are a group of enzymes, this, however, is an extensive and expensive process as HBOCs must be derived from raw hemoglobin, the HBOCs then also work as a group of enzymes rather than a fewer amount increasing the complexity and the biocompatibility of these HBOCs in the blood.
2. The use of enzymes is also critical as they are difficult to obtain, many biotech companies have jumpstarted the sale and production of enzymes, but the risk of versatility is posed and prolongs the process.
3. Many solutions are also not targeted toward neutralizing the effect but preventing it completely, making it more difficult to obtain results. Using enzymes to inverse this effect, allows results to quickly be seen as well as tangible.

Introducing My Solution

My approach is to design new enzymes that can neutralize the effects of H2O2 and Fe(III) to ultimately increase the shelf life of red blood cells. This is done using the Rosetta3 Algorithm and further developed with A2A adenosine receptors.

Phase 1: De Novo Enzyme Design with Rosetta3

Flavoenzymes are oxidoreductases that will mitigate the formation of Iron (III) as raw hemoglobin will break down into Iron (III) and Hydrogen Peroxide. The catalyze will then allow the Hydrogen Peroxide to separate into Hydrogen and Oxygen so that the toxic Hydrogen Peroxide’s effects are prevented.

My solution employs the Rosetta3 algorithm to design two enzymes that can reverse the effects of oxidation stress: An enzyme that breaks H2O2 into Water and Oxygen and an enzyme that changes Fe (III) into Fe (II). To do this, I will use a technique called De Novo Enzyme design. This is the process of designing new enzymes that aren’t based on parent enzymes but instead, creating new enzymes from scratch.

There are 4 main steps in this process:

Check out this paper for the full explanation

1. Choice of the catalytic mechanism and theoretical enzyme (theozyme)

The first step is to determine the reaction transition states and find the optimal structure for the catalyst. The optimal structure includes finding a set of idealized active site descriptions that consist of disembodied side chains and backbone functional groups. They then surround the transition state for the optimal catalyst which is derived from quantum chemistry saddle point calculations.

Once the theoretical enzyme (theozyme) has been defined, I can convert these terms into a geometric constrain file (cstfile) so that the Rosetta model can understand its input.

2. Identifying sites in the scaffold where our theozyme can be placed

I take the theozyme and place it into an existing protein structure using the RosettaMatch module. The model’s inputs are:

Protein structure from the cstfile
List of protein scaffolds from my library

RosettaMatch will run every scaffold in the list (once) and determine if the theozyme and this scaffold are compatible. The output will be the number of matches found.

In order to determine what is a match, the theozyme is inserted into the scaffold. This means that amino acid side chains of theozyme have been placed into the backbone of the scaffold. Afterward, the ligand (a molecule that binds to the enzyme — in this case, Fe(III) and H2O2) is placed into the cavity of the scaffold (without clashing with protein backbones).

In running RosettaMatch, the user needs to prepare each scaffold and decide which algorithm to use for each side chain of the theozyme:

Preparing scaffold for matching:
Usually, I would only need to use one subset of scaffold residues during the matching process because not all residues are equally likely to form a binding site (which is what connects the enzyme to the ligand). In addition, the computational costs of increasing the number of residues that I consider for the matcher.

Choosing a matching algorithm for the theozyme interaction:
There are two different matching algorithms: classic matching and secondary matching. These algorithms are used to build rotamers (position amino acids) at each part of the scaffold active-site positions for every interaction of the theozyme.

3. Designing the Found Sites

After a scaffold match is found, the optimal amino acid residues for other positions on the scaffold are determined in order to build an active site that will match the ligand (in this case, it will be Fe(III) and H2O2). This is done through 4 stages:

1. Determining which residues to design and which to repack

Rosetta divides the residues into 5 groups based on the distance between the alpha carbon and the ligand heavy atom. Only residues within a distance of 8 Angstrom will be set to designable.

2. Optimizing the catalytic interactions

This involves a gradient-based minimization of the theozyme’s structure before designing a protein. All active sites that are not part of the theozyme are mutated to alanine.

3. Cycles of sequence design/minimization (with catalytic constraints if specified)

This is where the actual sequence design happens. It employs a Monte Carlo algorithm which finds lower energy sequences for the non-catalytic residues (the part that doesn’t accelerate the reaction). After the sequence design, the outputted structure is minimized.

4. Unrestrained fixed sequence rotamer pack minimization

The previous stage produces thousands of designs. It is also recommended to redesign every theozyme a few times. From here, one must determine which structures are worth experimentally analyzing. Rosetta has a built in function to determine the best ligand score. It refers to metrics such as the number of hydrogen bonds present and the number of unsatisfied polar atoms.

4. Evaluating results

Since there are tons of possible outputs that are model-produced after stage 3, I need to perform an evaluation and ranking of our designed enzymes to determine the most optimal one. In order to select the right design, the model takes into consideration the following:

1. The design needs to be active
2. The ligand needs to have a good score (ex., binding energy)
3. The catalytic residues have to be in a competent conformation (i.e., the shape of the enzyme based on the amino acid sequences)
4. The active site must be pre-organized

Phase 2: Production of the Enzyme

Enzymes can be produced by taking A2A adenosine receptor proteins, which are present on the surface of most cells, and reacting them with copper-containing chemical groups that catalyze the Diels-Alder reaction. The Diels-Alder reaction is the cyclic transition that turns the diene protein into a subjectively reactive enzyme, like flavoenzymes.

The adenosine A2A receptor is a G-protein-coupled receptor that is important for controlling myocardial oxygen consumption, coronary blood flow, and CNS neurotransmitters.

The Diels-Alder reaction is the reaction between a conjugated diene and an alkene (dienophile) to form unsaturated six-membered rings.

The Method: Extracorporeal Enzyme Replacement Therapy

Extracorporeal Enzyme Replacement Therapy (EERT) is a method of using IV infusion to distribute an enzyme evenly in a patient’s body; by pumping the blood out and running it through a dialyzer before pumping the blood back into the patient.

EERT can be periodically used on donated bags of blood to analyze the amount of enzyme replacement needed and distribute the amount needed evenly throughout the blood before pumping it back into the bag. If the EERT process is repeated every 24 hours with the new enzymes then it can be enough to simulate the 25-hour period of new blood cells passing through the spleen in a human body.

This Process Outlined Above:

1. The blood from the transfusion bag is sucked out
2. The blood transfers through a spectrophotometer to determine the number of enzymes in the blood
3. AI calculates how much of each new enzyme is needed
4. The EERT puts the number of enzymes needed for the blood that passed through the spectrophotometer by the time it reaches the dialyzer
5. The blood is pumped back into the transfusion bag and put into refrigerated storage.

This method then allows immediate resorption of the enzyme into the blood, in order to ensure the collected blood sample does not allow the blood to break down once stored enzyme flavoenzymes and catalyze.

By adding these enzymes, the reactivity of the raw hemoglobin can then be neutralized allowing the blood once extracted from the body to be stored for much longer.

Economics Incentives and Overall Impact

Last year approximately 272 million dollars worth of blood was donated to the US. By increasing the shelf life of blood, we will:

1. Decrease the total cost/amount of blood donations
2. Decrease the risk factor of blood transfusion.

Ultimately, My solution will enable a new future to prevent the blood shortage crisis.