Advances in sequencing technologies have accelerated the discovery of the genetic causation of diseases and new cell types. This revolution has also given rise to a new technique known as genetic barcoding where DNA barcodes are used to investigate cellular heterogeneity and diverse cell-material interactions. Recently, DNA barcoding technologies have been applied to generate barcoded cells and nanoparticles to investigate heterogeneous cell-nanoparticle interactions to boost the translational application of nanomedicine.

DNA barcodes are a sequence of nucleotide bases that provide a unique tag to the labeled cells. Theoretically, the number of unique barcodes that can be achieved with genetic barcoding can easily be in the millions (four to the power of the base-pair length of the random barcode). For instance, a barcode of eight bases will generate 4,096 unique sequences. In practice, the number is usually smaller as some bases are used to prevent steric blocking and some bases are randomized to minimize bias during PCR.

Genetic DNA barcoding has its limitations. Compared to competing optical barcoding methods, most genetic barcoding does not offer any spatial information. which is crucial to decipher cell-cell interactions in tissues. [Janiecbros/Getty Images]
Once formed, the DNA barcodes are packaged into viral vectors to be introduced into cells. However, nonviral methods using electroporation can also be used for barcode transfection. In some cases where the target number of barcoded labeled cells is exceedingly high, nonviral methods may be more suitable as viral construction only allows for compact barcode design of about 30 random bases due to the presence of viral genetic elements in the viral vectors.

The objective is to integrate a unique barcode into each cell, and when the cell divides, the barcode will be duplicated and passed on to the daughter cells. Typically, the word “clone” is used to describe cells derived from a single barcoded cell. These cells share the same barcode. After treatment, the cells can be sequenced, and the barcodes are bioinformatically processed for quality control metrics. The number of barcode reads is used as a proxy for the percentage of cells with that barcode.

With this technique, researchers can monitor the fitness of multiple clones at the same time such as to map the clonal fate of parental cells contributing to tumor formation and metastasis by barcode frequency. Although the concept of cellular genetic barcoding was developed 25 years ago, new technologies such as cas9-generated barcodes are continually being developed, thus enabling millions of cells to be tracked over developmental and evolutionary time scales and to record cellular features in response to stimuli, including nanomedicine.1

However, the genetic DNA barcoding technique has its limitations.2 Compared to competing optical barcoding methods, most genetic barcoding does not offer any spatial information. which is crucial to decipher cell-cell interactions in tissues. In 2017, Frieda et al., introduced a technique named memory by engineered mutagenesis with optical in situ readout (MEMOIR) that combined multiplexed fluorescence in situ hybridization with CRISPR-cas9 generated barcodes.3

Recently, Simonson and co-workers developed a novel tyramide-conjugated DNA barcode technique with higher signal amplification while avoiding conjugating to DNA barcodes that can result in antibody dysfunction. Using their method, the team showed that they were able to make use of commercially available antibodies without the additional need for purification to spatially image biomarkers in classic Hodgkin lymphoma.4 Such techniques may be further integrated to understand spatial distribution of nanoparticles in cells.

Additionally, as the efficacy of this method still relies on viral transduction, it is crucial to optimize the multiplicity of infection (MOI) and the number of cells transduced. When these two parameters are optimized, it can avoid problems such as “repeat use” where a DNA barcode is used for cells that are genetically different, “multiple integration” where a cell is labeled with multiple barcodes and “multiple labeling” where multiple barcodes labeled identical cells.

Boehnke et al., lately made use of barcoded cell lines to discover cell and nanoparticle features to boost nanomedicine delivery.5 The team pooled and plated close to 500 barcoded cancer cell lines in a single well and screened their interactions against a range of nanoparticle formulations. It was found that the core composition of nanoparticles significantly affected their interactions with cells. Additionally, through genomic analyses, the team discovered that the gene SLC46A3 could be a predictive biomarker of cell-nanoparticle interaction as SLC46A3 negatively regulates liposomal nanoparticle uptake.

DNA barcoding for screening of lipid nanoparticles

Lipid nanoparticles have shown great promise for drug delivery as witnessed in their use for the SARS-CoV-2 vaccine. While thousands of nanoparticles can be generated each time with chemical modifications, it remains laborious to screen them first using cell culture in vitro before selecting a smaller number of candidates for in vivo testing. Importantly, in vitro efficacy can also be a poor predictor of in vivo efficacy, whole body biodistribution, and toxicity. To solve this challenge, in 2017, Dahlman et al., first demonstrated the idea of using DNA barcoded lipid nanoparticles for high throughput in vivo screening.6

DNA barcodes can be synthesized as described earlier. They were then formulated separately for each type of nanoparticle (i.e., barcode 1 to nanoparticle 1, and barcode N to nanoparticle N) before being pooled together for in vivo testing after checking that the barcodes did not affect delivery and that the various nanoparticles do not aggregate.

Using this method, James Dahlman, PhD, associate professor at the Georgia Institute of Technology, and co-workers identified chemical modifications such as tail length and molecular weight of polyethylene that affected lipid nanoparticle delivery to organs such as the liver and lungs. The best lipid nanoparticles were then chosen to deliver siRNA for transient inhibition of the protein, factor 7, in vivo. Recently, the Dahlman lab took a step further tapping on DNA barcoding to develop a method that they termed species-agnostic nanoparticle delivery screening (SANDS).7

SANDS technology tags different lipid nanoparticle formulations containing a reporter mRNA with a unique DNA barcode. When the nanoparticle is successfully delivered into preclinical models in vivo, the successfully transfected cells (aVHH+) can be sequenced to identify the specific nanoparticle formulation and how their features enable improved transfection efficiency. [Dahlman lab at Georgia Institute of Technology]
Each of their lipid nanoparticles carried a functional mRNA encoding a reporter and a DNA barcode. In total, they created 89 nanoparticle formulations tested in six murine models (some with humanized and primatized livers).

Based on their SANDS technology, the team identified a species-dependent response to lipid nanoparticles, including mRNA translation and endocytosis. Furthermore, a differential immune response was also proposed as a potential mechanism for strain-dependent delivery to hepatocytes. While the team acknowledged that the obtained results can be different for various nanomaterials and cell types of interest, they believe that SANDS is a good way to understand fundamental mechanisms for improving preclinical development of RNA delivery agents.

“Over the last few years, the lab has published screens that read out LNP functional delivery (FIND),8 work in any species (SANDS), and quantify delivery in single cells (SENT-seq).9 This progress was driven by hardworking Georgia Tech and Emory trainees. Our team now believes the next steps are to streamline the workflows so that more LNPs can be tested in a year,” said Dahlman.

“​SANDS allows scientists to screen in any desired preclinical model. This can be used for scientific questions or clinical translation. In one unpublished example, we injected barcoded LNPs into several genetic knockouts. By comparing delivery in mice with genes X, Y, or Z deleted, we could quickly evaluate how X, Y, and Z expression influenced delivery in vivo. In a clinical application, we envision screens in clinically relevant models, including disease models, and understanding how species influences delivery readouts,” Dahlman added.


  1. Kebschull, J. M., & Zador, A. M. (2018). Cellular barcoding: lineage tracing, screening and beyond. In Nature Methods (Vol. 15, Issue 11, pp. 871–879). Nature Publishing Group.
  2. Serrano, A., Berthelet, J., Naik, S. H., & Merino, D. (2022). Mastering the use of cellular barcoding to explore cancer heterogeneity. In Nature Reviews Cancer. Nature Research.
  3. Frieda, K. L., Linton, J. M., Hormoz, S., Choi, J., Chow, K. H. K., Singer, Z. S., Budde, M. W., Elowitz, M. B., & Cai, L. (2017). Synthetic recording and in situ readout of lineage information in single cells. Nature, 541(7635), 107–111.
  4. Simonson, P. D., Valencia, I., & Patel, S. S. (2022). Tyramide-conjugated DNA barcodes enable signal amplification for multiparametric CODEX imaging. Communications Biology, 5(1).
  5. Boehnke, N., Straehla, J. P., Safford, H. C., Kocak, M., Rees, M. G., Ronan, M., Rosenberg, D., Adelmann, C. H., Chivukula, R. R., Nabar, N., Berger, A. G., Lamson, N. G., Cheah, J. H., Li, H., Roth, J. A., Koehler, A. N., & Hammond, P. T. (2022). Massively parallel pooled screening reveals genomic determinants of nanoparticle delivery. Science, 377(6604).
  6. Dahlman, J. E., Kauffman, K. J., Xing, Y., Shaw, T. E., Mir, F. F., Dlott, C. C., Langer, R., Anderson, D. G., & Wang, E. T. (2017). Barcoded nanoparticles for high throughput in vivo discovery of targeted therapeutics. Proceedings of the National Academy of Sciences of the United States of America, 114(8), 2060–2065.
  7. Sago, C. D., Lokugamage, M. P., Paunovska, K., Vanover, D. A., Monaco, C. M., Shah, N. N., Castro, M. G., Anderson, S. E., Rudoltz, T. G., Lando, G. N., Tiwari, P. M., Kirschman, J. L., Willett, N., Jang, Y. C., Santangelo, P. J., Bryksin, A. v., & Dahlman, J. E. (2018). High-throughput in vivo screen of functional mRNA delivery identifies nanoparticles for endothelial cell gene editing. Proceedings of the National Academy of Sciences of the United States of America, 115(42), E9944–E9952.
  8. Ni, H., Hatit, M. Z. C., Zhao, K., Loughrey, D., Lokugamage, M. P., Peck, H. E., Cid, A. del, Muralidharan, A., Kim, Y. T., Santangelo, P. J., & Dahlman, J. E. (2022). Piperazine-derived lipid nanoparticles deliver mRNA to immune cells in vivo. Nature Communications, 13(1).
  9. Dobrowolski, C., Paunovska, K., Schrader Echeverri, E., Loughrey, D., da Silva Sanchez, A. J., Ni, H., Hatit, M. Z. C., Lokugamage, M. P., Kuzminich, Y., Peck, H. E., Santangelo, P. J., & Dahlman, J. E. (2022). Nanoparticle single-cell multiomic readouts reveal that cell heterogeneity influences lipid nanoparticle-mediated messenger RNA delivery. Nature Nanotechnology, 17(8), 871–879.