Dendrimers escape sensing by the human complement defence system

Dendrimers are synthetic nanostructures that grow like trees displaying surface motifs typically spaced less than 1 nm from each other. This surface motif arrangement evades the human complement defence system and opens new avenues for design of safer stealth nanomedicines and biomedical devices.

Like Comment
Read the paper

The human immune system is equipped with a battery of molecular sensors that detects exogenous and endogenous danger signals. Essentially, this is done through sensing of molecular patterns. For instance, the surfaces of many pathogens display repeating structural features (e.g., unique sugar motifs) at the nanometer-scale spacing arrangement that attract pattern-recognition molecules belonging to the complement defence system (a component of the innate immune system). The complement pattern-recognition molecules include C1q, mannan-binding lectin (MBL) and ficolins, which typically recognise patterns with a periodicity of 2–15 nm1,2 (this translates to a distancing arrangement of at least 5000 times thinner than the thickness of a typical sheet of paper). C1q is an archetypal multimeric pattern-recognition molecule that senses anionic clusters, among other targets, whereas MBL and ficolins sense carbohydrate (e.g., mannose) and acetylated pathogen-associated molecular patterns, respectively. Pattern sensing by complement pattern-recognition molecules leads to elimination of danger through complement opsonisation (this aids macrophage recognition of opsonised invaders for destruction) and direct lysis (e.g., through the formation and insertion of the complement membrane-attack complex into the pathogen walls). Surface pattern spacing and orientation display is also a double-edge sword. For example, some virulent pathogens display patterns of sialic acid molecules, which attract a complement regulator molecule (factor H). This protects the pathogen against complement attack and improves pathogen survival inside the host.

Since, the nanometer-scale spacing arrangement of surface patterns seems essential for the binding of complement pattern-recognition molecules1,2, we questioned whether complement activation can be avoided through a surface pattern display below the nanometer-scale spacing arrangement [i.e., in Angstrom-Scale Spacing Arrangement (ASSA)]. If this hypothesis holds true, then this could potentially open new strategies for design and engineering of complement-evading nanomedicines and biomaterials through molecular surface patterning. Thus, we turned our attention to dendrimers to test the ASSA hypothesis.

What are dendrimers?

Dendrimers are synthetic uniform branched architectures (Figure 1) with dimensions far smaller than viruses. A dendrimer has a core, an inner shell, and an outer shell, where the outer shell displays precise numbers of surface end-terminal motifs of different functionalities (e.g., carboxyl, pyrrolidone, hydroxyl and amine). Dendrimer sizes are controlled in a geometrical manner, which in turn is associated with an exponential increase in the number of surface end-terminal motifs. For instance, a second-generation dendrimer (2.9 nm in diameter) has 16 end-terminal motifs, whereas third- (3.6 nm in diameter) and fourth- (4.5 nm in diameter) generations display 32 and 64 end-terminal motifs, respectively. Dendrimeric surface motifs are typically spaced less than 1 nm from each other. Thus, dendrimers are unique uniform materials for testing the role of surface ASSA in complement sensing and activation.

Fig. 1 | A schematic representation of a fourth generation spherical poly(amido amine) dendrimer with pyrrolidone end-terminal functionality. There are 64 pyrrolidone groups on the surface. Considering lack of sufficient space near the dendrimer surface, it is likely that a few of the end-terminal motifs could back-fold and project inward.

What did we find out?

We tested our hypothesis with a library of different size dendrimers (3–6 nm in diameter) bearing identical core and shell structures, but with chemically different end-terminal motifs (carboxyl, pyrrolidone and amine functionalities). We showed all dendrimer types escaped sensing by C1q and MBL. Unlike many virulent pathogens, dendrimers even evaded factor H binding. When we studied complement activation in human plasma, pyrrolidone- and carboxylic acid-terminated dendrimers did not trigger any of the three complement pathways (classical, lectin and alternative). Furthermore, these dendrimers did not inhibit the function of the complement system. So, we were pleased to see that the surface ASSA concept was delivering its promise.

Next, we asked the question as to whether dendrimers could still escape complement if they form larger structures with similar sizes to viruses. This is an important consideration, since some non-specific plasma proteins could deposit on such nanoparticles, and these proteins could trigger complement activation3. Thus, we incorporated a planar hydrophobic molecule into dendrimers. This yielded dendrimeric nanoparticles with sizes spanning from 10 to around 100 nm. Again, these dendrimeric nanoparticles escaped complement sensing and no complement activation occurred. The outer shell of these dendrimeric nanoparticles is still composed of dendrimers, where the dendrimeric surface ASSA seems to orchestrate complement evasion.

But there is always a surprise!

While pyrrolidone- and carboxyl-terminated dendrimers always escaped complement sensing, a surprising finding was complement activation by amine-terminated dendrimers. Here, complement activation was strictly through the MBL arm of the lectin pathway, despite lack of dendrimer sensing by MBL. We initially thought this mode of lectin pathway activation could be due to direct activation of the MBL-associated serine proteases. This hypothesis proved wrong. Then we considered that this activation could arise from dendrimer binding to some non-specific blood proteins. But, which protein? The inspiration came from an earlier work by James Arnold and colleagues4 since they identified immunoglobulin M (IgM) glycoforms that bind to MBL. In a follow up paper Stahl’s group showed MBL binding to IgM activates the complement lectin pathway in vitro and in vivo5. Could IgM be the answer?

 The risk paid off!

It was race against time (our funding was nearly ending). We decided to put all our eggs into the same basket and try the IgM hypothesis. The risk paid off! We found that amine-terminated dendrimers predominantly hitchhike on a subset of natural IgM glycoforms causing IgM to trigger lectin pathway activation. Lessons learned; never take a free ride, especially on IgM!

 What are the implications of our findings?

  • Intravenously injected dendrimers can accumulate at pathological sites with leaky vasculatures, and dendrimers can carry therapeutic and diagnostic cargos to such diseased locations6. Thus, complement evasion by dendrimeric medicines is advantageous in conditions where local complement activation is problematic and promotes disease progression (e.g., solid tumours, macular degeneration, rheumatoid arthritis, and atherosclerotic plaques). Indeed, many nanomedicines today trigger local complement activation at diseased sites, raising concerns7.

  • Considering the complement-evading nature of dendrimers (particularly pyrrolidone- and carboxylic acid-terminated dendrimers), dendrimers could potentially be employed to functionalise complement-activating surfaces (e.g., drug nanocarriers, implants and biomedical devices), where deposition of a monomolecular layer of multifunctional dendrimers could overcome complement activation. Depending on dendrimer properties, the dendrimer monolayer may further act as a sensor (e.g., through its luminescent or magnetic properties) and a drug-eluting platform. Such plug-and-play strategies with dendrimers could open new avenues for engineering of complement-safe multifunctional nanomedicines and biomedical devices.

  • Considering our proposed ASSA concept in complement evasion, one might envisage the operation of a similar complement escape mechanism in some virulent pathogens. Thus, it might be possible that some pathogens express surface clusters of carboxyl and hydroxyl motifs in ASSA to avoid sensing by complement pattern-recognition molecules.

We are currently designing a range of complement-evading dendrimeric nanomedicines, theranostics and platforms for wide applications in medicine, particularly for combating atherosclerosis. We are also expanding our studies to better understand the role of pattern periodicity and dendrimeric architectural arrangements in innate immune responses.


In memory of Robert B. Sim; a complement pioneer.

Link to Article: Wu, L-P., Ficker, M., Christensen, J.B., Simberg, D., Trohopoulos, P.N., & Moghimi, S.M. Dendrimer end-terminal motif-dependent evasion of human complement and complement activation through IgM hitchhiking. Nature Communications 12: 4858 (2021).



  1. Sim, R. B. & Wallis, R. Immune attack on nanoparticles. Nat. Nanotechnol. 6: 80–81 (2011).
  2. Gjelstrup, L. C. et al. The role of nanometer-scaled ligand patterns in polyvalent binding by large mannan-binding lectin oligomers. J. Immunol. 188: 1292–1306 (2012).
  3. Vu, V. P. et al. Immunoglobulin deposition on biomolecule corona determines complement opsonization efficiency of preclinical and clinical nanoparticles. Nat. Nanotechnol. 14: 260–268 (2019).
  4. Arnold, J. N. et al. Human serum IgM glycosylation. Identification of glycoforms that can bind to mannan-binding lectin. J. Biol. Chem. 280: 29080–29087 (2005).
  5. McMullen, M. E. et al. Mannose-binding lectin binds IgM to activate the lectin complement pathway in vitro and in vivo. Immunobiol.  211: 759–766 (2006).
  6. Wu, L.-P., Ficker, M., Chistensen, J. B., Trohopoulos, P. N. & Moghimi, S. M. Dendrimers in medicine: therapeutic concepts and pharmaceutical challenges. Bioconjug. Chem. 26: 1198–1211 (2015).
  7. Moghimi, S. M., Simberg, D., Papini, E. & Farhangrazi, Z. S. Complement activation by drug carriers and particulate pharmaceuticals: principles, challenges and opportunities. Adv. Drug Deliv. Rev. 157: 83–97 (2020).

Seyed Moein Moghimi

Professor of Nanomedicine, Newcastle University