Challenges and strategies of spatial transcriptomics

Another challenge for ISS protocols is ensuring efficiency in the RNA-to-complementary DNA (cDNA) conversion step by reverse transcription (RT), which is considered inefficient and variable. One way to solve this issue is to eliminate the RT step. The method published in Science in 2018, named Spatially Resolved Transcript Amplicon Readout Mapping (STARmap), did precisely that —replacing the RT step by employing a pair of primer and padlock probes that hybridize to the same RNA followed by enzymatic amplification to produce a DNA nanoball (amplicon). Consequently, errors in data signal generation are also avoided as the probes only bind when perfect matches occur and are consequently stripped away following imaging². Hydrogel-tissue chemistry employed in the technique allowed the DNA amplicons to be embedded into the 3D tissue scaffold. STARmap has avoided the efficiency hurdle of cDNA conversion and achieved highly accurate and efficient simultaneous mapping of 160 to 1020 genes in mouse brain sections. Further, replacing the RT step in the STARmap method greatly improved specificity (by reducing the non-specific hybridization of single probes) and the efficiency of transcript capture.

To reap the benefits of ST in high-throughput patient tumour screening, ST protocols must be successfully integrated into clinical diagnostic workflows. The challenge in doing this is automating protocols to reduce the time taken to perform protocols in busy clinical laboratories. The research article by Maino et al., tackles this challenge by describing an automated ST method that can be performed in clinical diagnostic laboratories. Here, the ISS technique was adapted to an automated microfluidic system. The developed system involved a programmable system to deliver reagents in a temperature- and flow-controlled manner, which supported ISS protocols comprising of hybridization and ligation of gene-specific padlock probes and rolling circle amplification of probes and barcode sequencing using sequencing-by-ligation (SBL) chemistry. Although the automated technique demonstrated the same performance as the manual ISS protocols, the automation provided supplementary benefits such as reduced hands-on time (3h 12 mins compared to 4h 38 mins) and increased assay speed (10h compared to 18h 37 mins)³. Certain assay parameters of the automated method were not as ideal as the manual technique, such as the higher consumption of enzymes (1.5x higher than the manual ISS protocol), the benefits of time-saving and lower consumption of oligonucleotide (2.5x reduction) make the automated ISS technique very useful for routine clinical analysis³.

A common issue with ST methods is the high levels of background signals arising from autofluorescence in tissue samples, which can complicate signal acquisition and analysis. Imaging of human brain samples, particularly those of aging brains, is hampered by strong autofluorescence in multiple channels caused by lipid-containing residues of lipofuscin. This problem is compounded by the fact that human brains are much larger than those from other animal models often used to characterize ST methods. To reduce the problem of autofluorescence, ST methods have relied on additional computational methods, which are time-consuming. The Hybridization-based ISS(HybISS) ST method process employs a simple autofluorescence quenching step that circumvents the need for additional microscopic strategies or computational tools. This autofluorescence quenching step only adds five minutes to the protocol, and one-time treatment lasts well for image cycles⁵. The HybISS, together with the autofluorescence method, reduces non-specific background signals in different channels and achieves a clear separation of signals. This has been critical in distinguishing gene transcripts across entire human brain sections.

Human samples can often have poor RNA quality. This is the case for gut-associated tissue, which contains digestive enzymes that can rapidly degrade RNA, which can produce misleading transcriptomic data.  Researchers have addressed this issue by modifying a commercially available spatial transcriptomics platform, 10X Genomics’ Visium technology. In the past, one of the criteria for using Visium technology has been the integrity of RNA, defined by the RNA Integrity Number (RIN). The standard Visium protocol recommends having a RIN ≥ 7. The modified protocol, developed by Mirzazadeh et al., is RNA-Rescue Spatial Transcriptomics (RRST), with three main modifications. The protocol included formalin fixation instead of methanol and an additional baking step to prevent detachment and reinforce tissue section adhesion. The last modification was removing the crosslink-reversal step, further preventing RNA degradation. Employing the RRST method achieved higher detection rates for different regions of tissues with moderate to low RNA than the standard Visium protocol⁷.

1.          Chen, X., et al. Efficient in situ barcode sequencing using padlock probe-based BaristaSeq. Nucleic Acids Res46, (2018).

2.          Wang, X. et al. Three-dimensional intact-tissue sequencing of single-cell transcriptional states. Science (1979)361, (2018).

3.          Maïno, N. et al. A microfluidic platform towards automated multiplexed in situ sequencing. Sci Rep9, (2019).

4.          Eng, C. H. L. et al. Transcriptome-scale super-resolved imaging in tissues by RNA seqFISH+. Nature568, (2019).

5.          Gyllborg, D. et al. Hybridization-based in situ sequencing (HybISS) for spatially resolved transcriptomics in human and mouse brain tissue. Nucleic Acids Res48, (2020).

6.          Kim, Y. et al. Highly Multiplexed Spatially Resolved Proteomic and Transcriptional Profiling of the Glioblastoma Microenvironment Using Archived Formalin-Fixed Paraffin-Embedded Specimens. Mod Pathol36, (2023).

7.          Mirzazadeh, R. et al. Spatially resolved transcriptomic profiling of degraded and challenging fresh frozen samples. Nat Commun14, (2023).