Research paper
Isolation of microarray-quality RNA from primary human cells after intracellular immunostaining and fluorescence-activated cell sorting

https://doi.org/10.1016/j.jim.2013.02.003Get rights and content

Abstract

Microarrays have made it possible to perform high-throughput, genome-wide analyses of RNA expression from an extremely wide range of sources. This technology relies on the ability to obtain RNA of sufficient quantity and quality for this type of application. While there are means to circumvent limitations in the former, recovery of RNA suitable for microarray analysis still represents a major issue when working with some biological samples, particularly those treated with and preserved in nucleic acid-modifying organic reagents. In the present report we describe a procedure for the isolation of RNA suitable for microarray analysis from cells purified by fluorescence-activated cell sorting after fixation, permeabilization and intracellular staining with fluorochrome-conjugated antibodies. We show that – although the RNA isolated from these samples presented some degradation – it performed remarkably well in microarray analysis. The method we describe here makes it available to genome-wide expression profiling a variety of biological samples that so far were confined to single-gene analysis.

Graphical abstract

Highlights

► We describe a method to isolate microarray-quality RNA from cells purified by FACS. ► Cells were sorted after fixation, permeabilization and intracellular antibody staining ► Despite some degradation, recovered RNA performed remarkably well in microarray. ► It allows gene expression profiling of samples so far limited to single-gene study.

Introduction

Transcriptomics examines the expression levels of messenger RNA (mRNA) in a given cell population, using high-throughput techniques based on DNA microarray technology. The combination of transcriptomics and computational analysis has significantly advanced our knowledge of the processes governing the function of many cells, tissues and organisms in health and disease. The recent completion of several mammalian genome sequences provides the foundation to decoding the way in which genomes are translated into the functions of living organisms. The key intermediate is the transcriptome, which consists of all the individual RNA molecules (transcripts) produced by the cells. Recently, the analysis of > 20 million sequences has led to a comprehensive and detailed view of the mouse transcriptome (Schena et al., 1995). At the same time, the advent of microarrays made it possible to measure the RNA expression of thousands of genes in parallel and in a single assay (Lockhart et al., 1996, Shoemaker et al., 2001) allowing a refined high-throughput mapping of the transcriptional activity of a given genome (Bertone et al., 2004, Carninci et al., 2005).

The success of microarray analysis is predicated on several factors. Fragmentation and chemical modifications influence the overall quality of the RNA sample, as well as its template activity. RNA fragmentation is a function of time and temperature of sample storage. Unless properly inhibited, nearly ubiquitous RNase enzymes digest RNA molecules into shorter fragments that might compromise downstream applications (Auer et al., 2003). In the past, RNA integrity was evaluated using agarose gel electrophoresis stained with ethidium bromide, which typically produces two major bands comprising the 28S and 18S ribosomal RNA (rRNA) species (Sambrook et al., 1989). A ratio of 28S/18S ≥ 2.0 indicates RNA of high quality. However, this method is highly subjective and data cannot be processed digitally. The introduction of the bioanalyzer a decade ago by Agilent Technologies has allowed the separation of nucleic acid and protein samples under fully automated, digitized and reproducible conditions. This instrument employs microfluidics technology to perform eletrophoretic separations of tiny amounts of RNA samples in the channels of microfabricated chips according to their size and then detected by laser-induced fluorescence. Moreover, digitized information deriving from a large collection of RNA sample profiles recorded with an Agilent bioanalyzer was employed to extract an algorithm that describes RNA integrity in a user-independent, automated and reliable manner that allows the calculation of an RNA integrity number (RIN) (Schroeder et al., 2006). This value classifies RNA based on a numbering system from 1 to 10, with 1 being the most degraded and 10 being the most intact. Although the RIN cannot predict whether a given sample will work in any given assay, in general samples with RINs greater that 7–8 are expected to perform well in most applications, while samples with RINs below 7 require extra validation studies.

The presence and the extent of chemical modifications also contribute to the overall quality of the RNA sample. Nucleic acid-modifying organic agents (such as formaldehyde) routinely used to preserve biological specimens introduce mono-methylol groups (CH2OH) in all four bases at rates varying from 4% to 40% (Masuda et al., 1999). In addition, some adenine residues undergo dimerization through methylene bridging (Masuda et al., 1999). The extent of chemical modifications is dependent on the length of time in the presence of formaldehyde (Masuda et al., 1999, Chung et al., 2008). A number of studies have successfully carried out microarray studies using RNA recovered from formalin-fixed, paraffin-embedded samples (Li et al., 2006, Haque et al., 2007, Linton et al., 2008, Conway et al., 2009, Budczies et al., 2011, Xie et al., 2011).

An additional factor playing a role in the success of microarray analysis is the purity of the cellular source of RNA to be analyzed. Although this is not an issue when working with established cell lines, it is particularly relevant when working with primary specimens. Fluorescence activated cell sorting (FACS) allows to enrich to near homogeneity a specific cell type from a mixed population based. The differential expression of one or more biomarkers (for which specific monoclonal antibodies are available) can be exploited to discern the cell type of interest from all others, and to isolate it at purities often > 95–99%. On the other hand, the recovery of high quality RNA from FACS-purified cells poses some challenges. Indeed, the RNA can be partially degraded due to mechanical damage as well as to inefficient inactivation of RNases during cell sorting (Diez et al., 1999). Moreover, the use of fixative agents (such as formaldehyde) introduces chemical modifications that affect RNA performance in cDNA synthesis (Diez et al., 1999).

Here, we have taken an important and significant step further, and we report a method for the recovery of microarray-quality RNA from primary human cells purified to near homogeneity after fixation, permeabilization, intracellular staining with fluorochrome-conjugated antibodies, and FACS sorting. This method expands the application of microarray analysis to cells isolated from mixed populations (oligo- and polyclonal cell lines, cultured primary cells, clinical specimens) based upon differential expression of intracellular markers.

Section snippets

Generation of latently infected cells

For the studies described below we used cells from 4 different HIV-1 negative, healthy volunteers (4 biological replicates). The specimens used for this study were unmarked and could not be identified directly or through identifiers linked to the subjects, by any investigator on this study. As such, the Office of Research Subjects IRB at the University of Maryland, Baltimore has determined that studies involving these samples do not qualify as Human Subject Research, and do not require IRB

Results and discussion

We have established an in vitro model to generate primary human CD4+ T cells latently infected with HIV-1 (Marini et al., 2008). Our model yields a cell culture containing quiescent cells, both latently infected and uninfected. In order to gain a deeper understanding of the basic biology of HIV-1 latency, we sought to perform whole-genome expression profiling of latently infected CD4+ T cells in comparison to their uninfected counterparts. Currently, there are no known cellular biomarkers that

Competing interests

The authors declare no competing interests.

Acknowledgments

This work was supported by National Institutes of Health grant AI084711 and by Bill & Melinda Gates Foundation Grand Challenges Explorations grant OPP1035926 (F.R.). L.M. was also supported by NIH-NCI grant P30CA006973. This paper is subject to the NIH Public Access Policy.

References (22)

  • C. Diez et al.

    Isolation of full-size mRNA from cells sorted by flow cytometry

    J. Biochem. Biophys. Methods

    (1999)
  • H. Auer et al.

    Chipping away at the chip bias: RNA degradation in microarray analysis

    Nat. Genet.

    (2003)
  • P. Bertone et al.

    Global identification of human transcribed sequences with genome tiling arrays

    Science

    (2004)
  • J. Budczies et al.

    Genome-wide gene expression profiling of formalin-fixed paraffin-embedded breast cancer core biopsies using microarrays

    J. Histochem. Cytochem.

    (2011)
  • P. Carninci et al.

    The transcriptional landscape of the mammalian genome

    Science

    (2005)
  • J.Y. Chung et al.

    Factors in tissue handling and processing that impact RNA obtained from formalin-fixed, paraffin-embedded tissue

    J. Histochem. Cytochem.

    (2008)
  • C. Conway et al.

    Gene expression profiling of paraffin-embedded primary melanoma using the DASL assay identifies increased osteopontin expression as predictive of reduced relapse-free survival

    Clin. Cancer Res.

    (2009)
  • G. Fedorowicz et al.

    Microarray analysis of RNA extracted from formalin-fixed, paraffin-embedded and matched fresh-frozen ovarian adenocarcinomas

    BMC Med. Genomics

    (2009)
  • T. Haque et al.

    Gene expression profiling from formalin-fixed paraffin-embedded tumors of pediatric glioblastoma

    Clin. Cancer Res.

    (2007)
  • P. Kiewe et al.

    Prediction of qualitative outcome of oligonucleotide microarray hybridization by measurement of RNA integrity using the 2100 Bioanalyzer Capillary Electrophoresis system

    Ann. Hematol.

    (2009)
  • H.R. Li et al.

    Two-dimensional transcriptome profiling: identification of messenger RNA isoform signatures in prostate cancer from archived paraffin-embedded cancer specimens

    Cancer Res.

    (2006)
  • Cited by (10)

    • Isolation of intact RNA from murine CD4 <sup>+</sup> T cells after intracellular cytokine staining and fluorescence-activated cell sorting

      2018, Journal of Immunological Methods
      Citation Excerpt :

      Transcriptomic studies performed using RNA isolated by these methods showed that while fragmentation and modifications of isolated RNA is a concern, there is high correlation of transcriptome profiles between fresh frozen and FFPE samples (Hedegaard et al., 2014). However, when one of those commercial methods was applied to perform transcriptomic studies of primary human CD4+ T cells infected with HIV-1, the RNA isolated from these cells showed degradation that likely occurred during the ICS procedure (Iglesias-Ussel et al., 2013). More recent reports have proposed modifications in the buffers used for intracellular staining in addition to the use of modified RNA isolation protocols to isolate intact RNA from fixed and permeabilized cells.

    • High salt buffer improves integrity of RNA after fluorescence-activated cell sorting of intracellular labeled cells

      2014, Journal of Biotechnology
      Citation Excerpt :

      The single stranded nature of RNA molecules renders them sensitive to degradation, and crosslinking resulting from fixation has well known detrimental effects on RNA integrity and often precludes downstream analyzes such as gene arrays and quantitative real-time PCR (QPCR) (Auer et al., 2003; Fleige and Pfaffl, 2006). Several attempts have been made to improve protocols in order to isolate RNA from sorted cells with acceptable quality, however, RNA integrity values remain problematically low (Diez et al., 1999; Nishimoto et al., 2007; Iglesias-Ussel et al., 2013). Different RNA-preserving fixation solutions has also been developed, but these are often costly and not compatible with antibody labeling (Zaitoun et al., 2010).

    View all citing articles on Scopus
    View full text