Beyond the Liver: How Lipid Nanoparticle Conjugation Is Unlocking Cell and Tissue Targeting
- Michael Nguyen
- Apr 15
- 9 min read
Lipid nanoparticles have earned their place as the delivery platform of choice for RNA therapeutics. The COVID-19 vaccines proved they could work at global scale, and a growing pipeline of siRNA, mRNA, and gene editing therapies has validated their versatility. But there is a problem the field has been grappling with quietly for years: most LNPs, when administered systemically, end up in the liver.
For hepatic indications, that's fine. For everything else, such as T cells, tumor microenvironments, lungs, the central nervous system, it's a fundamental limitation. The liver's fenestrated endothelium and abundance of apolipoprotein E simply make it the path of least resistance for most nanoparticle formulations.
Targeted LNP conjugation is changing that paradigm. By decorating the surface of lipid nanoparticles with ligands that recognize specific cell-surface receptors, researchers are beginning to direct RNA payloads with a precision that was unthinkable a decade ago. This post explores how that's done, what chemistries make it possible, and where the field is heading.

The Targeting Problem, Why It Matters
Untargeted LNPs rely primarily on passive accumulation and endogenous protein adsorption — particularly ApoE — to reach hepatocytes after intravenous administration. This works well for liver-targeted gene silencing, as demonstrated by patisiran (Onpattro), the first FDA-approved RNA interference therapy. But for extrahepatic targets, passive delivery is inefficient, and in many cases insufficient.
T cells are a perfect illustration of the challenge. As the central effectors of adaptive immunity, T cells are among the most therapeutically compelling targets in oncology and autoimmunity. Engineering T cells with CAR constructs or immunomodulatory RNA payloads could transform how we treat cancer, autoimmune disease, and viral infections. The problem is that T cells are notoriously resistant to LNP uptake — they lack the endocytic machinery that makes hepatocytes and macrophages such efficient nanoparticle consumers.
Targeted conjugation addresses this by essentially giving the LNP a molecular address. Instead of relying on passive biodistribution, the nanoparticle is programmed to bind a specific receptor on the cell surface, triggering receptor-mediated endocytosis and delivering its payload directly to the intended target.
Antibody-LNP Conjugates: The CD Antigen Playbook
The most intuitive approach to targeted delivery is to attach an antibody — or antibody fragment — to the LNP surface, directed against a cell-type-specific surface marker. The CD (cluster of differentiation) antigen family has proven particularly useful here, offering a rich library of cell-type-specific surface proteins that can serve as molecular zip codes.
Anti-CD3 for Pan-T Cell Targeting
CD3 is expressed on the surface of virtually all mature T cells as part of the T cell receptor complex, making anti-CD3 one of the most widely studied targeting ligands for T cell-directed delivery. Researchers have conjugated anti-CD3 antibodies and anti-CD3 nanobodies (vHH fragments) to LNP surfaces to enable selective mRNA delivery to T cells both ex vivo and in vivo.
Kheirolomoom et al. demonstrated that anti-CD3-conjugated LNPs could deliver CAR-encoding mRNA directly to T cells in vivo in mice, generating functional CAR-T cells without the need for ex vivo cell manufacturing. This is a paradigm-shifting concept: rather than extracting patient T cells, engineering them in the lab, and reinfusing them — an expensive, time-consuming process — the therapy would be delivered intravenously and program T cells in their native environment.
Anti-CD4 and Anti-CD8 for T Cell Subset Targeting
Going one step further, anti-CD4 and anti-CD8 conjugation allows differential targeting of helper and cytotoxic T cell subsets respectively. This level of precision is particularly relevant for autoimmune applications, where selectively modulating CD4+ regulatory T cells without broadly activating CD8+ cytotoxic populations is critical for safety.
Anti-CD19 and Anti-CD22 for B Cell Targeting
For B cell malignancies and autoimmune conditions driven by aberrant B cell activity, anti-CD19 and anti-CD22 conjugated LNPs have shown promise as delivery vehicles for mRNA and siRNA payloads. CD19 in particular has been extensively validated as a therapeutic target through the success of CAR-T therapies like tisagenlecleucel, making it a well-characterized anchor for LNP targeting.
Anti-CD33 and Anti-CD123 for AML
In acute myeloid leukemia (AML), anti-CD33 and anti-CD123 conjugated LNPs are being explored to deliver apoptotic RNA payloads selectively to leukemic blasts while sparing normal hematopoietic progenitors. The selectivity advantage here is not just therapeutic — it's potentially the difference between a tolerable and an intolerable toxicity profile.
Beyond Antibodies: The Ligand Toolkit
Full-length antibodies are powerful targeting ligands, but they come with practical challenges — their large size can reduce LNP stability, increase immunogenicity, and complicate manufacturing. The field has responded by developing a diverse toolkit of alternative targeting moieties.

Nanobodies (vHH Fragments)
Single-domain antibody fragments derived from camelid heavy-chain antibodies, nanobodies offer antibody-like specificity in a fraction of the molecular footprint. Their small size (~15 kDa vs. ~150 kDa for a full IgG) makes them attractive for LNP surface conjugation, they create less steric interference with the lipid bilayer and are more amenable to site-specific conjugation chemistries. Anti-CD3 and anti-EGFR nanobodies conjugated to LNPs have shown strong preclinical targeting performance and are gaining traction as preferred ligands for next-generation targeted delivery.
Aptamers
Nucleic acid aptamers are short, structured oligonucleotides selected for high-affinity binding to specific targets, and offer another antibody-free approach to targeted LNP delivery. The anti-PSMA aptamer A10, for example, has been conjugated to LNP surfaces for targeted delivery to prostate cancer cells expressing prostate-specific membrane antigen. Aptamers are chemically synthesized, highly stable when modified, and can be conjugated site-specifically, making them well-suited to scalable manufacturing.
Peptides
Shorter still, targeting peptides derived from receptor-binding domains of natural ligands can be incorporated directly into lipid conjugates or attached to PEG chains on the LNP surface. RGD peptides targeting αvβ3 integrin, which is overexpressed on tumor vasculature and many cancer cells, have been widely studied. iRGD, a tumor-penetrating peptide variant, has shown particular promise for improving both tumor targeting and deep tissue penetration beyond the vascular compartment.
Small Molecule Ligands
For some targets, small molecule ligands offer the simplest path to targeted delivery. Folate receptor is overexpressed on many epithelial cancers, and folate-conjugated LNPs have been studied for tumor-targeted delivery for over two decades. GalNAc (N-acetylgalactosamine), which binds the asialoglycoprotein receptor (ASGPR) highly expressed on hepatocytes, has become the gold standard for hepatocyte-targeted delivery of oligonucleotide therapeutics — demonstrating that even for liver targeting, active targeting can dramatically improve potency over passive accumulation.
Conjugation Chemistry: How the Ligand Gets Attached
The targeting ligand is only as good as the chemistry that attaches it to the LNP surface. Poor conjugation can result in ligand loss during storage, random orientation that occludes the binding site, or linker instability that releases the ligand prematurely. The choice of conjugation chemistry is therefore not a minor technical detail but is central to the performance of the final construct.
Maleimide-Thiol Coupling (Classic Workhorse)
The most widely used conjugation strategy in LNP targeting involves reacting a maleimide-functionalized lipid on the LNP surface with a thiol group on the targeting ligand, typically introduced via reduction of a disulfide bond or site-specific incorporation of a cysteine residue. The reaction is fast, efficient, and well-characterized, and maleimide-PEG-lipids are commercially available for straightforward incorporation into standard LNP formulations.
The limitation is hydrolytic instability where maleimide-thioether bonds can undergo retro-Michael addition under physiological conditions, gradually releasing the ligand in circulation.
This has driven interest in more stable alternatives.
Click Chemistry: SPAAC and IEDDA (Modern Standard)
Bioorthogonal click chemistries have emerged as the preferred approach for precision conjugation in modern targeted LNP design. Two reactions in particular have become dominant:
Strain-Promoted Azide-Alkyne Cycloaddition (SPAAC, Azide-DBCO) — often called copper-free click chemistry — reacts a cyclooctyne-modified lipid with an azide-functionalized ligand under mild, biocompatible conditions with no metal catalyst required. The reaction is highly selective, proceeds at room temperature, and produces hydrolytically stable triazole linkages.
Inverse Electron Demand Diels-Alder (IEDDA) — involving tetrazine and trans-cyclooctene (TCO) reaction partners — is faster still, with rate constants several orders of magnitude higher than SPAAC. This is particularly advantageous when working with low-abundance surface ligands or when rapid, complete conjugation is needed at scale. IEDDA-based conjugation of antibodies and nanobodies to LNPs has produced highly homogeneous, orientation-controlled conjugates with excellent in vitro and in vivo targeting performance.
Site-Specific Conjugation: The Orientation Advantage
Regardless of the chemistry used, random conjugation is increasingly being replaced by site-specific approaches that ensure the targeting ligand is attached at a defined location — away from the binding interface. Strategies include:
Sortase-mediated ligation — using the bacterial enzyme sortase A to attach a ligand at a specific C-terminal LPXTG motif
Unnatural amino acid incorporation — genetically encoding non-canonical amino acids bearing bioorthogonal handles (azides, alkynes) at precise positions in the ligand sequence
C-terminal thiol engineering — introducing a single cysteine at the C-terminus of a nanobody or Fab fragment for site-specific maleimide or click conjugation
These approaches consistently outperform random conjugation in targeting efficiency because they maximize the proportion of ligands in an active, binding-competent orientation on the LNP surface.
Organ-Selective LNPs: Targeting Without a Ligand
A parallel and increasingly important strategy for extrahepatic delivery doesn't rely on surface conjugation at all — instead, it exploits the intrinsic sensitivity of LNP biodistribution to lipid composition. The Selective Organ Targeting (SORT) approach, pioneered by the Siegwart lab and published in Nature Nanotechnology in 2020, demonstrated that simply adding a fifth lipid component to standard four-component LNP formulations could redirect delivery from the liver to the lungs (with permanently cationic lipids), spleen (with anionic lipids), or maintain liver targeting (with additional ionizable lipids).
This compositional targeting approach is simpler to manufacture than surface-conjugated LNPs and has shown robust organ selectivity in multiple animal models. For applications where cell-type specificity within an organ is less critical than organ-level targeting — pulmonary delivery for cystic fibrosis or lung cancer, splenic delivery for systemic immune modulation — SORT-style formulation may offer the most practical path forward.
Clinical Translation: Where Are We Now?
Targeted LNP conjugates are still predominantly in preclinical development, but several programs are advancing toward and into clinical testing.
In Vivo CAR-T Generation
Perhaps the most clinically exciting application of targeted LNP conjugation is the in vivo generation of CAR-T cells, eliminating the need for ex vivo cell manufacturing entirely. Intellia Therapeutics, in collaboration with academic partners, has reported preclinical data on anti-CD3 and anti-CD5 targeted LNPs delivering CAR-encoding mRNA to T cells in vivo. Umoja Biopharma and several academic groups are advancing similar platforms. While no in vivo CAR-T LNP program has yet completed Phase I trials, the pace of preclinical development suggests clinical entry within the next 2–3 years.
GalNAc-LNP Conjugates for Hepatic Targeting
The clearest clinical validation of active targeting via LNP conjugation comes from GalNAc-based systems. Alnylam's GalNAc-siRNA conjugates — though not strictly LNP-based — have demonstrated the dramatic potency advantage that receptor-mediated hepatocyte targeting can provide, with subcutaneous dosing intervals extending to once-quarterly or even once-yearly administration. GalNAc incorporation into LNP formulations is being explored to extend these benefits to larger RNA payloads, including mRNA and gene editing constructs.
Tumor-Targeted LNPs
Multiple clinical-stage programs are evaluating antibody or peptide-targeted LNPs for oncology applications, including folate receptor-targeted LNPs for ovarian and endometrial cancer, and anti-EGFR conjugated LNPs for colorectal and head and neck cancers. Results to date have been mixed, reflecting the additional complexity of tumor targeting — heterogeneous antigen expression, poor tumor penetration, and immunogenic responses to antibody-coated particles remain active areas of optimization.
Key Considerations for Targeted LNP Design
For researchers and developers approaching targeted LNP conjugation for the first time, a few principles consistently determine success:
Ligand density matters — but more isn't always better. Over-decoration of the LNP surface can impair endosomal escape, reduce circulation half-life, and increase immunogenicity. Optimal ligand density needs to be empirically determined for each target-ligand combination. Often 0.1 - 2% of the overall formulation composition.
PEG architecture influences targeting efficiency. Standard PEGylation improves LNP stability and circulation, but the same PEG layer that prevents protein adsorption can also shield surface-conjugated ligands from receptor engagement. Cleavable or shorter PEG variants, or conjugation of ligands to the distal end of PEG chains, can help strike the right balance.
In vitro targeting data doesn't always translate in vivo. Serum protein adsorption rapidly remodels the LNP surface in biological fluids, forming a protein corona that can mask targeting ligands or alter biodistribution in ways that simple cell culture assays don't capture. Rigorous in vivo validation remains essential.
Manufacturing scalability should be considered early. Some conjugation strategies that perform beautifully at research scale become significant challenges at GMP manufacturing scale. Building scalability into conjugation chemistry selection from the outset can prevent costly reformulation later.
Looking Ahead
The convergence of improved conjugation chemistries, expanding ligand toolkits, and deeper understanding of LNP biophysics is rapidly maturing targeted delivery from a research curiosity into a clinically viable strategy. The in vivo CAR-T programs in particular represent a potential step-change in the economics and accessibility of cell therapy. If delivery can be accomplished with an off-the-shelf injectable rather than a patient-specific manufacturing process, the implications for cost, scalability, and global access are profound.
At Helix Biotech, we work closely with research teams navigating these exact challenges: from selecting the right conjugation chemistry for a given ligand and target, to formulating and characterizing targeted LNPs for preclinical studies. Whether you're exploring a well-validated CD antigen target or pioneering a novel receptor axis, getting the delivery right is foundational to everything that follows.
Interested in targeted LNP formulation for your program? Get in touch with the Helix Biotech team to discuss your project.

