Send the Enzyme With the Drug DNA That Encodes a Protein and the Machinery to Finish It

This application covers giving a patient two pieces of genetic code at once: one that instructs the body to make a therapeutic protein, and a second that instructs it to make an enzyme whose job is to chemically finish that protein once it exists. The finishing step is a post-translational modification, and the worked example is tyrosine sulfation of an anti-HIV protein that does not work properly without it. It was filed by The Wistar Institute in Philadelphia, whose vaccine group has spent more than a decade on biologics delivered as DNA rather than as purified protein.

The Problem

Many proteins only become fully functional after they have been chemically decorated. A post-translational modification attaches a functional group to a finished protein chain, and depending on which group and where, it can route the protein to a particular cellular pathway, change how long it survives in circulation, alter how it folds, or change what it binds. The specification’s position is that encoding a target protein on a DNA plasmid is now well established, but that encoding the protein is not the same as encoding a working protein.

The conventional route has its own version of the problem. Antibodies, erythropoietin and clotting factors are manufactured in mammalian cell lines, usually Chinese hamster ovary cells, and the specification notes that the product is heterogeneous in both the location and the extent of its modifications, with some of that variation reducing functionality. Manufacturing controls the cell line, not the chemistry inside it.

The case that drives the examples is an HIV protein called eCD4-Ig, a fusion of the CD4 receptor with an antibody constant region and a short peptide that mimics CCR5, the co-receptor HIV uses to enter cells. That mimic only works if particular tyrosine residues in it carry sulfate groups, because sulfation is what CCR5 itself relies on when HIV binds it. Sulfation is performed by a small family of sulfotransferase enzymes, and there is no reason to assume a muscle cell asked to make eCD4-Ig from an injected plasmid has enough of the right enzyme in the right compartment.

What This Invention Does

Claim 1 is a method of post-translationally modifying a synthetic protein in a subject. The step is a single administration of a composition containing two things: a first recombinant nucleic acid sequence encoding the synthetic protein, and a second recombinant nucleic acid sequence encoding a modifier protein, where the modifier protein carries out the modification inside the subject. Claim 13 covers the composition itself, the two sequences together, as a product rather than a method.

The dependent claims narrow it into the worked example. The modification is chosen from a list including sulfation, acetylation, N-linked and O-linked glycosylation, myristoylation, palmitoylation, SUMOylation, hydroxylation, methylation, ubiquitylation and oxidation. Where it is sulfation, the modifier protein is tyrosylprotein sulfotransferase 1 or 2. Where it is TPST2, the preferred version carries an IgE leader sequence, and the synthetic protein is an antigen, antibody or immunoadhesin, with eCD4-Ig named specifically.

The IgE leader is the practical detail that makes the scheme work. The target protein is aimed at the secretory pathway by its own leader sequence, so the enzyme has to reach the same compartment or it never meets its substrate. Of four enzyme constructs tested, TPST2 fitted with an IgE leader, which pulls it into the endoplasmic reticulum during translation, produced sulfation, while a version with the transmembrane motif deleted did not. Dose mattered in the other direction: a high dose of enzyme plasmid cut expression of the target protein by about two thirds, an effect traced to the plasmid load rather than to sulfation, since an enzyme that cannot sulfate the target caused a similar drop. One part enzyme plasmid to a thousand parts target plasmid gave saturating sulfation with no measurable loss of expression.

The functional payoff was measured in mice. Serum from animals given both plasmids neutralised HIV pseudotyped viruses more potently than serum from animals given the target plasmid alone. Against one isolate the concentration needed for half maximal neutralisation fell from about 1.09 to about 0.16 micrograms per millilitre, and across the panel sulfation improved potency against eight of the twelve HIV isolates compared, the largest single improvement being about tenfold.

Key Features

• Two coded sequences, one administration. The therapeutic protein and the enzyme that finishes it are both delivered as nucleic acid in a single composition, so the modification happens in the subject rather than in a factory.
• A broad menu of modifications. The claims list eleven modification types, from sulfation and glycosylation through to ubiquitylation, making the mechanism a platform rather than a single-product claim.
• Compartment targeting by leader sequence. Adding an IgE leader to TPST2 traffics the enzyme into the endoplasmic reticulum, which is where it has to be to reach a protein bound for secretion.
• A very low enzyme dose. One part enzyme plasmid to a thousand parts target plasmid was enough to saturate sulfation, while higher doses suppressed how much of the target protein was made.
• Measured gain in potency. Sulfated eCD4-Ig neutralised eight of the twelve isolates compared better than the unsulfated version, with roughly a sevenfold improvement against one and tenfold against another.
• Delivery by electroporation. The specification pairs the constructs with pulsed electric field delivery, the technique the applicant’s group has used across its DNA-encoded biologics work.
• Composition claims as well as method claims. Claim 13 covers the two-sequence composition on its own, which reaches a manufacturer of the product and not only a clinician using it.

Who Is Behind It

The Wistar Institute of Anatomy and Biology is an independent biomedical research institute in Philadelphia, founded in 1892 and the oldest of its kind in the United States. It is best known publicly for vaccine work, and its Vaccine and Immunotherapy Center is the group this filing comes from.

Three inventors are named: David Weiner, Megan Wise and Ziyang Xu. Weiner directs the centre and is one of the principal figures behind DNA-encoded monoclonal antibodies, the approach of injecting the genetic instructions for an antibody and letting the recipient’s own cells produce it, since carried into human trials. Wise and Xu were researchers in his laboratory at the time of the work.

The chain is set out plainly in the specification. This is a divisional under section 79B of the Patents Act 1990 of Australian application 2019284482, the Australian national phase of PCT/US2019/036470, filed on 11 June 2019. That international application claims priority from United States provisional application 62/683,344, filed on 11 June 2018.

Why It Matters

DNA-encoded biologics are attractive for the reasons that make manufactured biologics expensive: no bioreactor, no cold chain for a purified protein, and the possibility of sustained production from a single administration. The weakness has always been control. When the manufacturing happens inside a patient there is no purification step, no quality assay on the batch, and no opportunity to fix a protein that came out wrong. Co-delivering the modifying enzyme answers part of that, moving one piece of the finishing chemistry into the same design space as the protein.

The choice of eCD4-Ig is not incidental. It has been pursued as a candidate for durable HIV protection since a 2015 study showed sustained expression from an adeno-associated virus vector could protect macaques against repeated challenge, and its dependence on sulfation is a known vulnerability. Supplying the missing chemistry from a second plasmid at a thousandth of the dose is a narrow result with wide implications, because many broadly neutralising constructs and DNA-encoded vaccine antigens carry modifications of their own.

The filing pattern reflects an institute treating this as platform intellectual property. A divisional in Australia nearly eight years after the priority date, on a claim broad enough to cover any pairing of an encoded protein with an encoded modifying enzyme, is a position taken with licensing in mind rather than a single therapeutic.

Related Concepts

• Post-translational modification – the chemical finishing step this application moves inside the body.
• Tyrosine sulfation – the specific modification demonstrated in the examples.
• CCR5 – the HIV co-receptor whose sulfated peptide the therapeutic protein imitates.
• Electroporation – the delivery method used to get both plasmids into muscle tissue.
• Monoclonal antibody – the class of biologic the applicant’s DNA-encoded platform is built around.
• Plasmid – the vector carrying both the protein and the enzyme sequences.

---

AU 2026202134 was published in the Australian Official Journal of Patents on 9 April 2026 and is open for public inspection. Patent applications represent inventions that are sought to be protected and do not necessarily reflect commercially available products.