Walking back from office and rushing for our mandatory evening tea, we entered a huge street in an elite Hyderabad locality. This time the topic of discussion was that why has no one ever attempted to sequence the transcriptome of a single cell and perform proteomics on that very cell! CITE-seq did come up as a method but as we had correctly recalled, it deals with only a few protein molecules that are present on the surface of the cell not the entire proteome.
The real answer, or what we thought was the real answer, is you simply can’t!? The biochemistry of RNA extraction is not conducive for extraction of proteins and vice-versa. But given that this problem is so fundamental, we had an inkling that someone would have attempted to solve this. So we started looking up more methods.
Lo and behold, someone had taken a shot at this! Even though the method is not super popular, a group of Chinese researchers had published a protocol in 2023 named scSTAP (single-cell simultaneous transcriptome and proteome).
The following are interesting sections of the paper which back their method:
In the process of single-cell multi-omics analysis, how to coordinate the operation procedures of different omics, such as the choice of cell lysis agents, is also a problem that needs to be addressed. The commonly used cell lysis reagent was 0.1% RapiGest in single-cell proteome analysis and 0.2% Triton X-100 in single-cell transcriptome analysis. It is well known that most of the surfactants including Triton X-100 are MS incompatible due to their interference with MS detection. Therefore, we first tried to use 0.1% RapiGest to lyse the single oocyte samples. However, the result showed the oocyte samples could not be completely lysed using 0.1% RapiGest because of the barrier effect of the cellular zona pellucida on oocyte lysis (Figures S1A and S1B). To solve this problem, we developed an enzyme-assisted cell lysis method to achieve the complete lysis of single oocytes, with which 50 μg/mL Lys-C was used to digest the zona pellucida of the oocytes, after which 0.1% RapiGest was used to lyse the cell membrane of the oocytes. This enzyme-assisted method for cell complete lysis was combined with the PSS method for equal splitting of the cell lysate solution to achieve precisely quantitative division of transcriptome and proteome in single cells.
They seem to have engineered a work around for RNA lysis buffers which interfere with MS. Not only that, given that their method relies on extremely small volumes of the lysate they have had to engineer their way around more challenges:
The aim of this work is to develop a platform and workflow to achieve the simultaneous analysis of MS-based proteomes and full-length sequencing-based transcriptomes in the same single-cell individuals. At present, the single-cell proteome analysis technique plays a speed-determining role in the development of single-cell multi-omics analysis because the amount of proteins contained in a single cell is extremely small and proteins cannot be amplified, leading to greater challenges than single-cell transcriptome analysis. For the analysis of transcriptome and proteome in a single-cell individual, the primary problem to be solved is how to separate and transfer the very small amount of RNAs and proteins existing in the single cell for respective transcriptome sequencing and proteome analysis.
Based on our previously developed sequential operation droplet array (SODA) technique,34,35 which enables precise metering and manipulation of liquids in the nanoliter-to-picoliter range, we developed an approach called precise sample splitting (PSS) to achieve the quantitative division of single-cell samples in the nanoliter range for simultaneous transcriptome and proteome analysis. To realize such a straightforward multi-omics analysis strategy adopting the precise splitting of microsamples for separate multi-omics analysis, two prerequisites should be met. The first prerequisite is that the system should have the ability to precisely split single-cell lysates in the nanoliter range, since nanoliter-scale sample volumes were usually adopted in most of the reported single-cell proteome analysis systems to depress the excessive dilution and adsorption loss of ultra-trace proteins. Only when a system has such a quantitative sample-splitting capability can it precisely and reproducibly control the splitting volumes and ratios for each single-cell sample, so as to ensure the comparability between the datasets of transcriptome and proteome obtained from different single cells. The second prerequisite is that after the single cell is lysed and before the lysate is divided, the cellular components should be uniformly distributed in the single-cell lysate solution. Otherwise, even if sample splitting can be performed quantitatively, the uneven distribution of the single-cell components such as RNAs and proteins will cause inconsistency in the component content between the two aliquots. This will lead to inaccurate quantitative results, deteriorating the reliability of single-cell multi-omics analysis results and the comparability of datasets from different single-cell samples. Besides the two prerequisites, a higher requirement for the PSS-based single-cell multi-omics analysis is that the performance of the simultaneous two-omics analysis should not have a remarkable decrease in the identification depth compared with the single-omics analysis.
The paper was recently corrected because the authors had failed to provide their code, apologizing nearly two years after publication. Now this was interesting to say the least.
We will read and understand this paper further to figure out where this method still lacks and why is it not that popular.
An interesting tangent worth mentioning is from the modeling world. This method had claimed to reproduce CITE-seq level of surface proteome using transcriptomics and some transformer magic. More on this later.