Quantitative single-cell RNA-seq with unique molecular identifiers.

Single-cell RNA sequencing (RNA-seq) is a powerful tool to reveal cellular heterogeneity, discover new cell types and characterize tumor microevolution. However, losses in cDNA synthesis and bias in cDNA amplification lead to severe quantitative errors. We show that molecular labels--random sequences that label individual molecules--can nearly eliminate amplification noise, and that microfluidic sample preparation and optimized reagents produce a fivefold improvement in mRNA capture efficiency.

Characterization of the single-cell transcriptional landscape by highly multiplex RNA-seq.

Our understanding of the development and maintenance of tissues has been greatly aided by large-scale gene expression analysis. However, tissues are invariably complex, and expression analysis of a tissue confounds the true expression patterns of its constituent cell types. Here we describe a novel strategy to access such complex samples. Single-cell RNA-seq expression profiles were generated, and clustered to form a two-dimensional cell map onto which expression data were projected. The resulting cell map integrates three levels of organization: the whole population of cells, the functionally distinct subpopulations it contains, and the single cells themselves-all without need for known markers to classify cell types. The feasibility of the strategy was demonstrated by analyzing the transcriptomes of 85 single cells of two distinct types. We believe this strategy will enable the unbiased discovery and analysis of naturally occurring cell types during development, adult physiology, and disease.

Validation of whole genome amplification for analysis of the p53 tumor suppressor gene in limited amounts of tumor samples.

Personalized cancer treatment requires molecular characterization of individual tumor biopsies. These samples are frequently only available in limited quantities hampering genomic analysis. Several whole genome amplification (WGA) protocols have been developed with reported varying representation of genomic regions post amplification. In this study we investigate region dropout using a φ29 polymerase based WGA approach. DNA from 123 lung cancers specimens and corresponding normal tissue were used and evaluated by Sanger sequencing of the p53 exons 5-8. To enable comparative analysis of this scarce material, WGA samples were compared with unamplified material using a pooling strategy of the 123 samples. In addition, a more detailed analysis of exon 7 amplicons were performed followed by extensive cloning and Sanger sequencing. Interestingly, by comparing data from the pooled samples to the individually sequenced exon 7, we demonstrate that mutations are more easily recovered from WGA pools and this was also supported by simulations of different sequencing coverage. Overall this data indicate a limited random loss of genomic regions supporting the use of whole genome amplification for genomic analysis.

Expression profiling of signature gene sets with trinucleotide threading.

In recent years, studies have shown that expression profiling of carefully chosen intermediary gene sets, comprising approximately 10 to 100 genes, can convey the most relevant information compared to much more complex whole-genome studies. In this paper, we present a novel method suitable for expression profiling of moderate gene sets in a large number of samples. The assay implements the parallel amplification features of the trinucleotide threading technique (TnT), which encompasses linear transcript-based DNA thread formation in conjunction with exponential multiplexed thread amplification. The amplifications bestow the method with high sensitivity. The TnT procedure together with thread detection, relying on thread-specific primer extension followed by hybridization to universal tag arrays, allows for three distinction levels, thus offering high specificity. Additionally, the assay is easily automated and flexible. A gene set, comprising 18 protein epitope signature tags from the Swedish Human Protein Atlas program, was analyzed with the TnT-based approach and the data were compared with those generated by both real-time PCR and genome-wide cDNA arrays, with the highest correlation observed between TnT and real-time PCR. Taken together, expression profiling with trinucleotide threading represents a reliable approach for studies of intermediary gene sets.

Allelotyping by massively parallel pyrosequencing of SNP-carrying trinucleotide threads.

Here we present an approach for allelotyping combining the multiplexing features of the trinucleotide threading (TnT) method with pooling of genomic DNA and massively parallel pyrosequencing, enabling reliable allele frequency estimation in large cohorts. The approach offers several benefits as compared to array-based methods and allows undertaking highly complex studies without compromising accuracy, while keeping the workload to a minimum. This proof-of-concept study involves formation of trinucleotide threads, targeting a total of 147 single-nucleotide polymorphisms (SNPs) related to obesity and cancer, for multiplex amplification and allele extraction from a pool of 462 genomes, followed by massively parallel pyrosequencing. Approximately 177k reads were approved, identified, and assigned to SNP-carrying threads rendering representative allele frequencies in the cohort.

Topographical transcriptome mapping of the mouse medial ganglionic eminence by spatially resolved RNA-seq.

BACKGROUND: Cortical interneurons originating from the medial ganglionic eminence, MGE, are among the most diverse cells within the CNS. Different pools of proliferating progenitor cells are thought to exist in the ventricular zone of the MGE, but whether the underlying subventricular and mantle regions of the MGE are spatially patterned has not yet been addressed. Here, we combined laser-capture microdissection and multiplex RNA-sequencing to map the transcriptome of MGE cells at a spatial resolution of 50 µm. RESULTS: Distinct groups of progenitor cells showing different stages of interneuron maturation are identified and topographically mapped based on their genome-wide transcriptional pattern. Although proliferating potential decreased rather abruptly outside the ventricular zone, a ventro-lateral gradient of increasing migratory capacity was identified, revealing heterogeneous cell populations within this neurogenic structure. CONCLUSIONS: We demonstrate that spatially resolved RNA-seq is ideally suited for high resolution topographical mapping of genome-wide gene expression in heterogeneous anatomical structures such as the mammalian central nervous system.

Base preferences in non-templated nucleotide incorporation by MMLV-derived reverse transcriptases.

Reverse transcriptases derived from Moloney Murine Leukemia Virus (MMLV) have an intrinsic terminal transferase activity, which causes the addition of a few non-templated nucleotides at the 3' end of cDNA, with a preference for cytosine. This mechanism can be exploited to make the reverse transcriptase switch template from the RNA molecule to a secondary oligonucleotide during first-strand cDNA synthesis, and thereby to introduce arbitrary barcode or adaptor sequences in the cDNA. Because the mechanism is relatively efficient and occurs in a single reaction, it has recently found use in several protocols for single-cell RNA sequencing. However, the base preference of the terminal transferase activity is not known in detail, which may lead to inefficiencies in template switching when starting from tiny amounts of mRNA. Here, we used fully degenerate oligos to determine the exact base preference at the template switching site up to a distance of ten nucleotides. We found a strong preference for guanosine at the first non-templated nucleotide, with a greatly reduced bias at progressively more distant positions. Based on this result, and a number of careful optimizations, we report conditions for efficient template switching for cDNA amplification from single cells.

Targeted transcript profiling by sequencing.

In this work we present a targeted gene expression strategy employing trinucleotide threading (TnT) amplification and massive parallel sequencing. We have previously shown that TnT combined with array readout accurately monitors expression levels. However, with this detection strategy spurious products go undetected. Accordingly, we adapted the TnT protocol to massive parallel sequencing to acquire an unbiased view of the entire TnT-generated product population. In this manner we investigated the identity of undesired products, their extent at different oligonucleotide:RNA ratios and their effect on the expression levels. We demonstrate that TnT gene expression profiling with massive sequencing readout renders reliable expression data from as low as 3.5 ng of total RNA. Moreover, using 350 ng of total RNA results in only 0.7% to 1.1% undesired products. When lowering the amount of input material, the undesired product fraction increases but this does not influence the expression profiles.

RNA-seq analysis reveals different dynamics of differentiation of human dermis- and adipose-derived stromal stem cells.

BACKGROUND: Tissue regeneration and recovery in the adult body depends on self-renewal and differentiation of stem and progenitor cells. Mesenchymal stem cells (MSCs) that have the ability to differentiate into various cell types, have been isolated from the stromal fraction of virtually all tissues. However, little is known about the true identity of MSCs. MSC populations exhibit great tissue-, location- and patient-specific variation in gene expression and are heterogeneous in cell composition. METHODOLOGY/PRINCIPAL FINDINGS: Our aim was to analyze the dynamics of differentiation of two closely related stromal cell types, adipose tissue-derived MSCs (AdMSCs) and dermal fibroblasts (FBs) along adipogenic, osteogenic and chondrogenic lineages using multiplex RNA-seq technology. We found that undifferentiated donor-matched AdMSCs and FBs are distinct populations that stay different upon differentiation into adipocytes, osteoblasts and chondrocytes. The changes in lineage-specific gene expression occur early in differentiation and persist over time in both AdMSCs and FBs. Further, AdMSCs and FBs exhibit similar dynamics of adipogenic and osteogenic differentiation but different dynamics of chondrogenic differentiation. CONCLUSIONS/SIGNIFICANCE: Our findings suggest that stromal stem cells including AdMSCs and dermal FBs exploit different molecular mechanisms of differentiation to reach a common cell fate. The early mechanisms of differentiation are lineage-specific and are similar for adipogenic and osteogenic differentiation but are distinct for chondrogenic differentiation between AdMSCs and FBs.

Highly multiplexed and strand-specific single-cell RNA 5' end sequencing.

Single-cell analysis of gene expression is increasingly important for the analysis of complex tissues, including cancer, developing organs and adult stem cell niches. Here we present a detailed protocol for quantitative gene expression analysis in single cells, by the sequencing of mRNA 5' ends. In all, 96 cells are lysed, and their mRNA is converted to cDNA. By using a template-switching mechanism, a bar code and an upstream primer-binding sequence are introduced simultaneously with reverse transcription. All cDNA is pooled and then prepared for 5' end sequencing, including fragmentation, adapter ligation and PCR amplification. The chief advantage of this approach is the great reduction in cost and time, afforded by the early bar-coding strategy. Compared with previous methods, it is more suitable for large-scale quantitative analysis, as well as for the characterization of transcription start sites, but it is unsuitable for the detection of alternatively spliced transcripts. Sample preparation takes 3 d, and two sets of 96 cells can be prepared in parallel. Finally, the sequencing and data analysis can take an additional 4 d altogether.

Common variants in human CRC genes as low-risk alleles.

The genetic susceptibility to colorectal cancer (CRC) has been estimated to be around 35% and yet high-penetrance germline mutations found so far explain less than 5% of all cases. Much of the remaining variations could be due to the co-inheritance of multiple low penetrant variants. The identification of all the susceptibility alleles could have public health relevance in the near future. To test the hypothesis that what are considered polymorphisms in human CRC genes could constitute low-risk alleles, we selected eight common SNPs for a pilot association study in 1785 cases and 1722 controls. One SNP, rs3219489:G>C (MUTYH Q324H) seemed to confer an increased risk of rectal cancer in homozygous status (OR=1.52; CI=1.06-2.17). When the analysis was restricted to our 'super-controls', healthy individuals with no family history for cancer, also rs1799977:A>G (MLH1 I219V) was associated with an increased risk in both colon and rectum patients with an odds ratio of 1.28 (CI=1.02-1.60) and 1.34 (CI=1.05-1.72), respectively (under the dominant model); while 2 SNPs, rs1800932:A>G (MSH6 P92P) and rs459552:T>A (APC D1822V) seemed to confer a protective effect. The latter, in particular showed an odds ratio of 0.76 (CI=0.60-0.97) among colon patients and 0.73 (CI=0.56-0.95) among rectal patients. In conclusion, our study suggests that common variants in human CRC genes could constitute low-risk alleles.

Analysis of short tandem repeats by parallel DNA threading.

The majority of studies employing short tandem repeats (STRs) require investigation of several of these genetic markers. As such, we demonstrate the feasibility of the trinucleotide threading (TnT) approach for scalable analysis of STRs. The TnT method represents a parallel amplification alternative that addresses the obstacles associated with multiplex PCR. In this study, analysis of the STR fragments was performed with capillary gel electrophoresis; however, it should be possible to combine our approach with the massive 454 sequencing platform to considerably increase the number of targeted STRs.