New approaches for treatment of acute intermittent porphyria by enzyme substitution and gene therapy : evaluation in vitro and in vivo

Abstract

Acute intermittent porphyria (AIP) is a rare inborn metabolic error of heme synthesis in which the activity of the third enzyme, porphobilinogen deaminase (PBGD), is deficient. It is clinically characterized by acute, potentially life-threatening neurologic attacks that are precipitated by various drugs, reproductive hormones and other factors. During acute attacks, the porphyrin precursors 5-aminolevulinic acid (ALA) and porphobilinogen (PBG) accumulate and are excreted in urine. Current treatment based on heme replenishment and carbohydrate loading reduces the accumulation of ALA and PBG. However, the therapy is palliative and cannot prevent the acute attacks. The aim of this study was to evaluate the potential of two new therapeutic alternatives, i.e. enzyme substitution and gene therapy, first in vitro using different cell lines and subsequently in vivo using an AIP mouse model.

The first step in the development of gene therapy for AIP was based on non-viral vectors, investigating whether the PBGD enzyme could be expressed in vitro in PBGD-deficient cells to correct the enzyme deficiency as well as the resulting biochemical defect. Four vectors encoding the mouse and human PBGD enzyme were constructed. Expression of PBGD after transfection of the plasmids, condensed to polyethylenimine (PEI), was found to be dose and time dependent. When non-viral gene transfer was undertaken in PBGD-deficient fibroblasts, the enzyme deficiency was corrected. The presence of a biochemical defect was demonstrated by measuring the synthesis of the heme precursor, protoporphyrin, after addition of ALA or PBG to the system. The human PBGD-deficient cells synthesized 21-36 % of the amount of protoporphyrin synthesized by control cells. By increasing the enzyme activity over the level found in control cells the protoporphyrin synthesis increased to 127-152 % of normal, showing that the biochemical defect could be corrected. Similar results were obtained in mouse PBGD-deficient cells. Before starting the in vivo trials, the AIP mouse model was investigated during induction of heme synthesis by phenobarbital to find out in which tissues the surplus of ALA and PBG is formed and to study the excretion patterns of these porphyrin precursors. With a LC-MS method it was possible to detect high levels of ALA and PBG in the liver and smaller amounts in the kidney. These observations point to the liver as the primary target organ for the gene therapy. During a four-day induction of heme synthesis by phenobarbital, the levels of ALA and PBG in plasma and urine gradually increased.

The effect of enzyme substitution on the levels of ALA and PBG in plasma and urine were studied in the AIP mouse model. Administration of recombinant human PBGD (rhPBGD) intravenously or subcutaneously after a four-day phenobarbital administration period was shown to lower the PBG level in plasma in a dose-dependent manner, with maximal effect seen after 30 min and 2 hours, respectively. Injection of rhPBGD subcutaneously twice daily during a four-day phenobarbital induction period reduced urinary PBG excretion to 25 % of the levels found in PBGD-deficient mice given only phenobarbital. This demonstrates efficient removal of PBG in plasma and urine by enzyme replacement. In spite of PBG clearance, no reduction of ALA concentration was seen at any time. The next step towards gene therapy was to study if the PBGD enzyme could be expressed in vivo in the AIP mouse model with the liver as the target organ. Four non-viral vectors were evaluated using luciferase as reporter enzyme: naked DNA and DNA complexed to liposomes, PEI and PEI-galactose.

After tail-vein injection of the DNA complexes, the highest luciferase expression was found in the lung. When injected into the portal vein, the naked DNA showed considerably higher hepatic reporter gene expression, 100 µg of naked DNA having the highest hepatic luciferase expression 24 h after injection. When these vectors were used to deliver the PBGD cDNA into the AIP mouse model no enhancement of hepatic PBGD activity was detected. The strategy for in vivo gene therapy was therefore changed to the use of recombinant adenoviral vectors, as they are known to efficiently target the liver in rodents. By the use of adenovirus encoding luciferase (Ad-EGFPLuc) it was confirmed that the liver was the main target organ after intravenous administration. When the PBGD-deficient mice were injected with adenovirus encoding PBGD (Ad-PBGD) the hepatic PBGD activity increased in a dose- and time-dependent manner.

The effect of the increased hepatic PBGD activity on ALA and PBG accumulation was also studied during phenobarbital-induction of heme synthesis. No accumulation of ALA or PBG could be found in plasma, liver or kidney, demonstrating that gene delivery of PBGD to the liver can prevent the accumulation of the presumably toxic porphyrin precursors. In conclusion, these experiments provide in vivo proof of concept for enzyme substitution and gene therapy in AIP. Enzyme substitution was thus found to reduce the PBG level in plasma, while gene delivery targeting the liver could correct the metabolic defect and prevent the accumulation of both ALA and PBG. These findings suggest a potential for both therapies in AIP.