Protein Therapy for Epidermolysis Bullosa (Marinkovich 2)

What did this project achieve?

This funding allowed us to develop a robust manufacturing process for collagen VII (C7), the protein missing in recessive dystrophic epidermolysis bullosa (RDEB). We developed this process using an industry standard cell line (CHO cells), as well as current industry standard production and purification technology. Full length gene coding for C7 was inserted into these cells and a process of multiple rounds of cloning and selection was performed on these CHO cells to induce them to secrete higher levels of increasingly more stable and high quality C7 protein. C7 stability is important because it likely will determine how long lasting the protein will be after it is introduced into patient skin.

We tested stability of C7 purified from these clones by a method called circular dichroism which looks at the wavelengths of light which are absorbed.  When the C7 falls apart, we can tell by looking at these wavelengths of light. Using this method, we were able to show that the C7 falls apart only at a high temperature, which suggests its able to withstand external forces and still hold together. This implies good stability.

C7 acts as a glue to hold the skin together and part of the way it does this is to glue itself to two other basement membrane proteins collagen IV and laminin-332. We tested to make sure that our manufactured C7 stuck to  collagen IV properly. In this way, we were able to verify that the C7 protein we produced in the lab functioned normally to glue itself into the basement membrane.

We analyzed a protein called prolyl-4-hydroxylase which is an enzyme which chemically modifies C7 in a way that makes it more stable. In the CHO cells we used to make C7 for us, we selected for those cells that produced the best and the most C7 protein. Thus it was probably was not a coincidence that the very same cells that produced the most/best C7 also were the ones that produced the most prolyl-4-hydroxylase.

The clone producing the highest levels of stable C7 was selected, and robust C7 expression was obtained in the clone even after thirty passages of growth. When a cell behaves in this way, we call it a stabile cell line. This means it has the potential for robust C7 production during a two week period of growth in a big 20 liter plastic bag filled with cell culture medium which is gently heated and gently rocked back and forth, just like a baby in a heated blanket being rocked in a mother’s arms. We call this type of apparatus a WAVE reactor. We found that the cell line was able to produce lots of C7 during the WAVE reactor culture run.

The cell line production and purification technology was transferred to the University of California at Davis good manufacturing process (GMP) Facility. This is a facility which can produce clinical grade C7 suitable for a clinical trial for RDEB patients. A master cell bank of C7 CHO cells was produced in this facility and we are currently in the process of performing viral and other testing on this cell bank, according to FDA requirements.

After recombinant human C7 (rC7) is expressed in our CHO cell line, using a WAVE reactor we next purify it using a two step ion process, followed by sterile filtration. We developed the technology to make C7 97% pure with low contamination of CHO cell proteins and DNA. We found it was safe when injected into 6 mice at a dose of 1.33 μg/gram- with no evidence of skin irritation, erythema, or weight loss after 3 months of monitoring. The recombinant C7 was functional and potent. As little as 0.3 ug (which is a very small amount) of C7 protein injected into the skin of RDEB mice was detected in the basement membrane zone by our microscopic methods. Our efforts to evaluate another enzyme which modifies C7 called C-proteinase were limited due to the lack of availability of effective antibodies which could be used on tissue sections for microscopic visualizatiojn.

Another goal of this grant was to prepare for C7 protein therapy clinical trials. Based on these studies, we propose a Phase 1 clinical trial to determine whether injection of rC7 into wound edges is safe and has the possibility of reducing wound area size. For safety purposes in this early trial, we will inject rC7 over a small surface area of less than 2% total body area. Our stock solution of C7 is 0.4 mg/ml in sterile buffer, and we anticipate injecting up to 4 wound edges (each 20 cm circumference or less) and one small area of intact skin. Our goal is to determine whether rC7 is functional and localizes to the basement membrane after 2 months and whether chronic wounds are reduced in size over 3 months.

Functionality will be determined by biopsies that will be assayed using indirect immunofluorescence microscopy to demonstrate C7 incorporation and persistence in the BMZ, and with immunoelectron microscopy to demonstrate the increased anchoring fibril formation and localization of rC7 into these structures. Blood will be monitored for anti-C7 antibodies. The primary goal of this five patient trial is to determine whether therapeutic rC7 can be safely administered and demonstrated to localize to the BMZ of RDEB patient skin to improve patient wounds. This information will enable us to determine if injection of C7 protein could be a developed as a potential treatment for patients with RDEB.

One final step for us will be to perform a toxicology study on 24 mice. The mice will be separated into three groups each receiving a different concentration of rC7. The duration of the study will be 3 months with repeated skin injections. Once we complete these studies and obtain FDA approval of the above plan, we will begin our clinical trial of C7 protein gene therapy.