---
title: "Designing for Delivery: How Construct Architecture Changes Across mRNA, AAV and Lentiviral Applications"
description: "Good construct design starts with a deceptively simple question: What does the final molecule need the cell to do?"
image: https://blog.genewiz.com/hubfs/genewiz-blog/Blog%20-%20Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20Across%20mRNA,%20AAV,%20Lentivirus/Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20Across%20mRNA%2c%20AAV%2c%20Lentivirus%20Header%20Image.jpg
---

[![GENEWIZ](https://blog.genewiz.com/hubfs/GENEWIZ%20Brand%20Assets/GENEWIZ%20Logos/2025-04%20Current%20GENEWIZ%20Logos/GENEWIZ_Full-Color_Logo_PublicWeb_200x63.svg "GENEWIZ")](https://www.genewiz.com/)

- Close

 Search

[GENEWIZ Blog Homepage](https://blog.genewiz.com/)

# Designing for Delivery: How Construct Architecture Changes Across mRNA, AAV and Lentiviral Applications

[By GENEWIZ .](https://blog.genewiz.com/author/genewiz)

 November 20, 2026

 0 Comments

![construct design](https://blog.genewiz.com/hubfs/genewiz-blog/Blog%20-%20Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20Across%20mRNA,%20AAV,%20Lentivirus/Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20Across%20mRNA%2c%20AAV%2c%20Lentivirus%20Header%20Image.jpg)

Choosing mRNA, AAV or lentivirus is only the first decision.

The same protein-coding sequence can behave very differently across these platforms because expression is shaped by much more than the open reading frame. The cap, UTRs, promoter, poly(A) signal and other regulatory elements influence how much protein is produced, where it is expressed, and how long expression lasts. So good construct design starts with a deceptively simple question: What does the final molecule need the cell to do?

## mRNA: Optimize the Whole Molecule; Not Just the Coding Sequence

![Designing for Delivery How Construct Architecture Changes Across mRNA, AAV and Lentiviral Applications-1](https://blog.genewiz.com/hs-fs/hubfs/genewiz-blog/Blog%20-%20Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20Across%20mRNA%2c%20AAV%2c%20Lentivirus/Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20-%20Figure%201.png?width=2172&height=724&name=Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20-%20Figure%201.png)

A conventional linear therapeutic, mRNA typically contains a 5′ cap, 5′ UTR, coding sequence, 3′ UTR, and poly(A) tail. Each component can influence stability and protein production, making mRNA design a multi-variable optimization problem rather than simply a codon-optimization exercise.\[1\]

| Element | Key design consideration |
| --- | --- |
| 5′ cap | Stability and translation initiation |
| 5′ UTR | Ribosome recruitment, structure, upstream AUGs |
| Coding sequence | Codon usage, GC content, RNA structure |
| 3′ UTR | Transcript stability and translation |
| Poly(A) | Stability and translation efficiency |

Nucleoside chemistry adds another layer. N1-methylpseudouridine (m1Ψ), commonly used in therapeutic mRNA design, can alter innate immune recognition and support expression. Recent work also shows that m1Ψ can subtly influence translation fidelity in a sequence- and codon-context-dependent manner.\[1,2\]

The intended protein matters too. A secreted antibody requires an appropriate signal peptide, while an mRNA encoding a genome editor may instead be designed to generate sufficient protein for editing while limiting the duration of enzyme exposure.

## AAV: Every Base Pair Competes for Space

AAV creates a very different design problem: space is limited.

Recombinant AAV vectors are generally most efficiently packaged when the vector genome remains close to the wild-type size of approximately 4.7 kb. As genomes become oversized, packaging efficiency can decline, and incomplete or heterogeneous genomes become more common.\[3\]

That ~4.7 kb budget includes everything between the ITRs:

![Designing for Delivery How Construct Architecture Changes Across mRNA, AAV and Lentiviral Applications-2](https://blog.genewiz.com/hs-fs/hubfs/genewiz-blog/Blog%20-%20Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20Across%20mRNA%2c%20AAV%2c%20Lentivirus/Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20-%20Figure%202.png?width=2172&height=724&name=Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20-%20Figure%202.png)

The promoter therefore directly affects how much room remains for the transgene. A large promoter may provide strong or well-characterized expression but consume substantial space. A compact promoter can recover hundreds of base pairs, although expression strength and tissue specificity must also be considered. The same trade-off applies to introns, WPRE, and other regulatory elements: an element may improve expression, but its benefit has to justify the space it occupies.

Self-complementary AAV makes that trade-off even sharper. scAAV can accelerate expression in some settings by bypassing conventional second-strand synthesis, but usable capacity falls to approximately 2.2–2.4 kb.\[4\]

For genes that genuinely exceed a single-vector design, dual-AAV approaches can divide a payload between two vectors and reconstitute it in the target cell. This strategy now has clinical precedent. In April 2026, the FDA granted accelerated approval to OTARMENI™ for certain patients with OTOF-associated sensorineural hearing loss, making it the first FDA-approved dual-AAV vector-based gene therapy.\[5\]

## Don't overlook the ITRs

AAV ITRs are essential for vector production and are structurally prone to rearrangement during plasmid propagation. Engineered constructs may intentionally use modified ITR configurations, but unintended deletions can also occur. A 2025 Nucleic Acids Research study documented ITR instability in commonly used plasmids and showed that ITR integrity can influence subsequent vector production and quality.\[6\]

## Lentiviral Vectors: Design for What Happens After Integration

Modern self-inactivating lentiviral transfer vectors retain the cis-acting sequences required for vector production and gene transfer while removing viral coding genes from the transfer genome. After integration, the therapeutic cassette is typically driven by an internal promoter.\[7\]

![Designing for Delivery How Construct Architecture Changes Across mRNA, AAV and Lentiviral Applications-3](https://blog.genewiz.com/hs-fs/hubfs/genewiz-blog/Blog%20-%20Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20Across%20mRNA%2c%20AAV%2c%20Lentivirus/Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20-%20Figure%203.png?width=2172&height=724&name=Designing%20for%20Delivery%20How%20Construct%20Architecture%20Changes%20-%20Figure%203.png)

Key elements can include:

| Element | Role |
| --- | --- |
| Ψ | Packaging signal for vector RNA |
| RRE | Supports Rev-dependent export of unspliced vector RNA during production |
| cPPT/CTS | Commonly included to support efficient transduction |
| Internal promoter | Controls transgene expression after integration |
| Transgene | Encodes the therapeutic product |
| WPRE | Can enhance RNA processing and expression |
| SIN 3′ LTR | Reduces native LTR promoter/enhancer activity after integration |

Because the vector integrates, promoter choice can shape expression for the lifetime of the engineered cell and its progeny. And maximum expression is not always the goal.

In CAR-T cells, for example, receptor expression and signaling need to be balanced. Depending on the CAR architecture, excessive constitutive expression can contribute to tonic signaling and T-cell dysfunction. Promoter selection is one way to tune that behavior. \[8\] Hematopoietic stem-cell applications may need something different again: stable, lineage-appropriate expression rather than simply the strongest possible promoter.

Instead of asking: **Which promoter is the strongest?**

Ask: **How should this gene behave for the lifetime of the engineered cell?**

Lentiviral vectors can accommodate larger genomes than AAV, but no sharp packaging cutoff has been defined; increasing vector-genome size generally reduces functional titer and can complicate manufacturing.\[9\]

## One Gene, Three Design Problems

The coding sequence may be identical, but the engineering question changes with the platform:

**mRNA:** How efficiently can this RNA produce the intended protein, and how should its stability and duration be controlled?

**AAV:** Can the complete cassette fit while maintaining the expression pattern the application needs?

**Lentiviral vector:** How should the gene behave after stable integration into the engineered cell?

The delivery platform does more than carry the coding sequence - **it changes the rules for how the construct should be designed.**

## References

1. Metkar M, Pepin CS, Moore MJ. Tailor made: the art of therapeutic mRNA design. *Nat Rev Drug Discov.* 2024;23:67–83.
2. Monroe J, Eyler DE, Mitchell L, et al. N1-Methylpseudouridine and pseudouridine modifications modulate mRNA decoding during translation. *Nat Commun.* 2024;15:8119.
3. Bennett A, Mietzsch M, Agbandje-McKenna M. Understanding capsid assembly and genome packaging for adeno-associated viruses. *Future Virol.* 2017;12:283–297.
4. McCarty DM. Self-complementary AAV vectors; advances and applications. *Mol Ther.* 2008;16(10):1648–1656. doi:10.1038/mt.2008.171.
5. U.S. Food and Drug Administration. FDA approves first-ever gene therapy for treatment of genetic hearing loss. April 23, 2026.
6. Radukic MT, Le DT, Krassuski T, et al. Degradation and stable maintenance of adeno-associated virus inverted terminal repeats in *E. coli*. *Nucleic Acids Res.* 2025;53.
7. Poletti V, Mavilio F. Designing lentiviral vectors for gene therapy of genetic diseases. *Viruses.* 2021;13:1526.
8. Liu D, Lee K-Z. Optimizing viral transduction in immune cell therapy manufacturing: key process design considerations. *J Transl Med.* 2025;23:501.
9. Sweeney NP, Vink CA. The impact of lentiviral vector genome size and producer cell genomic to gag-pol mRNA ratios on packaging efficiency and titre. *Mol Ther Methods Clin Dev.* 2021;21:574–584.

## Daily Insight

## Recent Posts

- [next generation sequencing](https://blog.genewiz.com/tag/next-generation-sequencing)
- [gene synthesis](https://blog.genewiz.com/tag/gene-synthesis)

[Designing for Delivery: How Construct Architecture Changes Across mRNA, AAV and Lentiviral Applications](https://blog.genewiz.com/how-construct-architecture-changes-across-mrna-aav-and-lentiviral-applications)

[By GENEWIZ.](https://blog.genewiz.com/author/genewiz) November 20, 2026

- [next generation sequencing](https://blog.genewiz.com/tag/next-generation-sequencing)
- [gene synthesis](https://blog.genewiz.com/tag/gene-synthesis)

[Transient, Durable, or Stable? mRNA, AAV, and Lentiviral Vectors Support Different Therapeutic Applications](https://blog.genewiz.com/transient-durable-or-stable-mrna-aav-and-lentiviral-vectors-support-different-therapeutic-applications)

[By GENEWIZ.](https://blog.genewiz.com/author/genewiz) November 18, 2026

- [next generation sequencing](https://blog.genewiz.com/tag/next-generation-sequencing)
- [gene synthesis](https://blog.genewiz.com/tag/gene-synthesis)

[The Inaugural International AIntibody Challenge: Putting AI-Designed Antibodies to the Test](https://blog.genewiz.com/inaugural-international-aintibody-challenge-putting-ai-designed-antibodies-to-the-test)

[By Crystal Richardson.](https://blog.genewiz.com/author/crystal-richardson) November 11, 2026

- [next generation sequencing](https://blog.genewiz.com/tag/next-generation-sequencing)
- [gene synthesis](https://blog.genewiz.com/tag/gene-synthesis)

[From Cave Cyanobacteria to Living Building Materials: How Sequencing Supports Discovery in Extreme Environments](https://blog.genewiz.com/from-cave-cyanobacteria-to-living-building-materials-how-sequencing-supports-discovery-in-extreme-environments)

[By GENEWIZ.](https://blog.genewiz.com/author/genewiz) November 04, 2026

## Recent Post Tags

- [next generation sequencing](https://blog.genewiz.com/tag/next-generation-sequencing)
- [gene synthesis](https://blog.genewiz.com/tag/gene-synthesis)
- [rna sequencing](https://blog.genewiz.com/tag/rna-sequencing)
- [antibody discovery](https://blog.genewiz.com/tag/antibody-discovery)
- [proteomics](https://blog.genewiz.com/tag/proteomics)
- [sanger sequencing](https://blog.genewiz.com/tag/sanger-sequencing)
- [transcriptomics](https://blog.genewiz.com/tag/transcriptomics)
- [cell and gene therapy](https://blog.genewiz.com/tag/cell-and-gene-therapy)
- [olink](https://blog.genewiz.com/tag/olink)
- [Event](https://blog.genewiz.com/tag/event)
- [GENEWIZ Update](https://blog.genewiz.com/tag/genewiz-update)
- [aav-itr](https://blog.genewiz.com/tag/aav-itr)
- [antibody engineering](https://blog.genewiz.com/tag/antibody-engineering)
- [itr sequencing](https://blog.genewiz.com/tag/itr-sequencing)
- [mRNA](https://blog.genewiz.com/tag/mrna)
- [single-cell sequencing](https://blog.genewiz.com/tag/single-cell-sequencing)
- [aav](https://blog.genewiz.com/tag/aav)
- [aav plasmid prep](https://blog.genewiz.com/tag/aav-plasmid-prep)
- [genomics](https://blog.genewiz.com/tag/genomics)
- [grant winner](https://blog.genewiz.com/tag/grant-winner)
- [lentivirus](https://blog.genewiz.com/tag/lentivirus)
- [quantitative pcr (qPCR)](https://blog.genewiz.com/tag/quantitative-pcr-qpcr)
- [rna therapeutics](https://blog.genewiz.com/tag/rna-therapeutics)
- [whole plasmid sequencing](https://blog.genewiz.com/tag/whole-plasmid-sequencing)
- [NULISA](https://blog.genewiz.com/tag/nulisa)
- [PacBio](https://blog.genewiz.com/tag/pacbio)
- [aquaculture](https://blog.genewiz.com/tag/aquaculture)
- [bioengineering](https://blog.genewiz.com/tag/bioengineering)
- [bispecific antibodies](https://blog.genewiz.com/tag/bispecific-antibodies)
- [codon optimization](https://blog.genewiz.com/tag/codon-optimization)
- [customer spotlight](https://blog.genewiz.com/tag/customer-spotlight)
- [digital pcr (dPCR)](https://blog.genewiz.com/tag/digital-pcr-dpcr)
- [direct colony sequencing](https://blog.genewiz.com/tag/direct-colony-sequencing)
- [immunity](https://blog.genewiz.com/tag/immunity)
- [long-read sequencing](https://blog.genewiz.com/tag/long-read-sequencing)
- [machine learning](https://blog.genewiz.com/tag/machine-learning)
- [microbiology](https://blog.genewiz.com/tag/microbiology)
- [oxford nanopore sequencing (ont)](https://blog.genewiz.com/tag/oxford-nanopore-sequencing-ont)
- [pcr + sanger sequencing](https://blog.genewiz.com/tag/pcr-sanger-sequencing)
- [peptides](https://blog.genewiz.com/tag/peptides)
- [plasmid synthesis](https://blog.genewiz.com/tag/plasmid-synthesis)
- [sample submission guidelines](https://blog.genewiz.com/tag/sample-submission-guidelines)
- [single-cell rna-seq](https://blog.genewiz.com/tag/single-cell-rna-seq)
- [whole genome sequencing](https://blog.genewiz.com/tag/whole-genome-sequencing)