Reference Genes for Quantitative qPCR Analyses in Monocytes of Septic Patients

Background: Monocytes and macrophages are essential components of the innate and adaptive immune responses and play a critical role in sepsis. Sepsis is a life-threatening organ dysfunction associated with an unregulated host response to infection. About 20 million people develop sepsis annually, and up to 50% die. There is a lack of studies regarding human monocytes and sepsis. This study aimed to determine the most stable internal gene (s) to investigate gene expression in monocytes/macrophages of septic patients. Methods: The expression stability of fteen commonly used reference genes was analyzed by determining the comparative threshold cycle (Ct) values, using the BestKeeper, GeNorm, and NormFinder algorithms. Results: BestKeeper analysis revealed that the syntaxin 5 (STX5A) and hypoxanthine phosphoribosyltransferase 1 (HPRT1) genes were highly stable. GeNorm pointed out STX5A and phosphoglycerate kinase 1 (PGK1) as the most suitable combination whereas through NormFinder glyceraldehyde 3- phosphate dehydrogenase (GAPDH) and 14-3-3 zeta/delta protein (YWHAZ) was the most stable combination. All programs analysis discarded the use of heterogeneous nuclear ribonucleoprotein A/B (HNRNPAB). GeNorm and NormFinder indicated actin-beta (ACTB) as the least stable gene. Conclusions: The combined data indicated that STX5A, PGK1, GAPDH, and HPRT1 are highly suitable reference genes for qPCR analysis of septic patients monocytes. In the case of choosing one single reference gene, the results point out to STX5A (rst place by GeNorm and BestKeeper and third place by NormFinder). This study is the rst report on reference genes in monocytes/macrophages from septic patients.


Background
Monocytes and macrophages are the front line innate defense against pathogens [1,2] Monocytes are immature immune cells that circulate in the bloodstream, and they are locally recruited and become mature macrophages when there is tissue injury [3] Macrophages are plastic cells and can differentiate into pro-in ammatory (M1) or anti-in ammatory and tissue repair type (M2) according to the microenvironment [3,4]. Monocytes and macrophages play an essential role in immunomodulation that occurs during sepsis [5]. According to the most recent de nition (Sepsis-3; 2016), sepsis is a life-threatening organ dysfunction associated with an unregulated host response to infection. Septic shock is a shock subset that exhibits circulatory and metabolic dysfunctions with high mortality risk [6,7]. About 20 million people develop sepsis annually, and 30%-50% die. Sepsis is still a public health problem [8].
Monocytes and macrophages have a set of receptors known as Pattern Recognition Receptors (PRR), which rapidly initiate the in ammatory response upon tissue damage or in the presence of pathogens. The toll-like receptors (TLRs) are a subfamily of PRRs and are the major cellular components required for pathogen-associated molecular patterns (PAMPs) recognition and in ammatory response initiation. Among TLRs, TLR4 mediates lipopolysaccharide (LPS) recognition in the cell wall of gram-negative bacteria. Lipoteichoic acid and glycan peptides found in gram-positive bacteria activate TLR2. The activation of the TLR2 triggers an in ammatory response in sepsis [9,10], and systemic activation of the immune response occurs due to the release of the PAMPs or DAMPs from microorganisms or tissue damage, respectively [9].
Several parameters ensure reliability and reproducibility in the quantitative gene expression measurements. The mentioned parameters are variations in initial sample volume, RNA integrity, cDNA synthesis e ciency, and differences in overall transcriptional activity in tissues or cells analyzed [17]. The most widely used approach to normalization of the interest gene expression is an inner control gene. The internal control gene would be the one that detection is constant and continued in all tissues. The internal control gene expression does not change in any experimental condition and treatment, allowing the normalization of the interest gene [17][18][19]. Usually, there are one or more genes constitutively expressed. Firstly, it is necessary to identify potential candidates and then to test their expression stability in the condition under investigation [17,19]. Every mRNA analysis requires stably expressed reference genes as control. Reproducibility and reliability of the qPCR results demand reference genes [19] Herein, we determined the most stable internal genes in monocytes/macrophages of septic patients. This is the rst report on reference genes in monocytes/macrophages freshly obtained from septic patients.

Literature search strategy
We used the National Center for Biotechnology Information (NCBI) public PubMed database (http/www.ncbi.nlm.nih.gov/pubmed) for article search. We de ned combinations using SEPSIS, SEPTIC SHOCK, MONOCYTE, MACROPHAGE, REFERENCE GENE, and GENE EXPRESSION.

Selection of samples
We selected samples from nine patients with a clinical diagnosis of sepsis admitted to the Intensive Care Unit (ICU) of the University Hospital of the University of Sao Paulo in the city of Sao Paulo, Brazil. We collected samples from patients during ICU hospitalization (n = 4) and immediately after ICU discharge (5), totalizing nine samples.
Samples from septic patients were collected whenever allowed by the clinicians, so only a limited number of patients could be used. The number of samples reached the recommendations for the GeNorm tool analysis [20].

Monocyte preparation
We collected 20 mL of blood in tubes containing EDTA from septic patients. Samples were centrifuged at 400 x g to obtain cells. We diluted the cells in phosphate buffer saline pH 7.4 and added Histopaque 1077 (Sigma Chemical Co, St Louis, MO, USA). We centrifuged the tubes at 400 x g and 4 °C for 30 minutes resulting in the isolation of mononuclear cells, which were cultured for one hour in RPMI-1640 medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (Sigma-Aldrich) at 37 °C in a 5% CO2 incubator. Following incubation, the monocytes adhered to the culture plate were isolated in Trizol® (Invitrogen, Carlsbad, California, EUA) and stored at -80 °C for further analysis.

Selection of candidates for reference genes
We rstly listed 15 reference genes previously used in studies with monocytes/macrophages from different sepsis experimental approaches mostly in bone marrow-derived cells and J7741 murine macrophage cell line (Table 1).
We obtained speci c primers for each gene using the National Center for Biotechnology Information (NCBI) public Gene Bank database (http/www.ncbi.nlm.nih.gov/genbank). We used the Primer-BLAST program (NCBI, Bethesda, Maryland, USA) to design the primers ( Table 2). The hybridization temperature of the primers was standardized for the experiment (60 °C). cDNA, and 1 µM forward and reverse primers (EXXTEND, Paulínia, São Paulo, Brazil); and 6.9 µL nuclease-free water, added to a 96-well plate. The cycling conditions followed the protocol of the SYBR Green manufacturer. We analyzed the melt curve to determine the speci city of the qPCR products and the absence of primer-dimer formation.

Bioinformatic tools used
The expression stabilities of the reference genes were determined using the BestKeeper (www.genequanti cation.de/bestkeeper.html, version 1), GeNorm (www.qbaseplus.com, version 3.2), and NormFinder (moma.dk/Norm nder-software, version 0.953) software. The BestKeeper software uses the coe cient of variance of gene expression to analyze which reference gene is the most stable one [21]. The GeNorm algorithm determines the stability by calculating the arithmetic mean of all pairwise variations of a particular reference gene relative to another one [19]. The NormFinder algorithm classi es the set of genes according to their expression stability in a given collection of samples and experimental design [17]. Expression pro le of candidate reference genes using the BestKeeper software

Results
As mentioned before, the BestKeeper software determines the stability of the candidate genes performing a pairwise comparative analysis based on crossing points, and the best stability values indicated by the lowest standard deviation and coe cient of variance [21]. The BestKeeper software identi ed the STX5A as the most stable and 18S as the less stable gene among 15 evaluated (Table 3) considering their standard deviation and coe cient of variance values.  The analysis indicated STX5A and PGK1 as the best reference genes based on the high stability and low combined variation (Fig. 2).
Expression pro le of the 15 reported reference genes using the NormFinder algorithm The NormFinder algorithm allows estimating the gene expression variation intergroup and intragroup. High stability of gene expression is indicated by a low stability value that is associated with the systematic error of each reference gene [17]. The best stability value is almost equal to zero. The ACTB and HNRNPAB genes exhibited higher expression stability value ( Table 4). The NormFinder analysis indicated the GAPDH as the most suitable single gene for qPCR normalization in sepsis and GAPDH and YWHAZ as the most stable combination. Ranking of the 15 reported reference genes as indicated by the three software tools According to the BestKeeper analysis, the expression of the STX5A, HPRT1, and PGK1 genes was more stable, with the lowest values of standard deviation and coe cient of variance. The NormFinder analysis indicated that a combination of GAPDH and YWHAZ has high expression stability. The GeNorm classi ed STX5A and PGK1 as the most top stable genes with a low combinatorial variation. Expressions of ACTB, HNRNPAB, and 18S genes were considered the least stable ones, according to the highest M value (GeNorm), coe cient of variance (BestKeeper), and stability value (NormFinder), respectively (Table 5).

Discussion
Only few studies reported results of gene expression in monocytes/macrophages in sepsis experimental approaches ( Table 1). The authors mostly performed in vitro studies using mice bone marrow-derived monocytes/macrophages [18,22], J774A1 murine macrophage cell line [23], monocytes from healthy donors stimulated with LPS [24] and monocytes from patients with relapsing-remitting multiple sclerosis [25]. It is notable the lack of study in monocytes/macrophages from septic patients. Fifteen different genes were used as reference genes to normalize qPCR results in monocytes/macrophages under the sepsis experimental conditions mentioned.
We then determined the most appropriate reference genes for studies in monocyte/macrophage freshly obtained from septic patients.
In response to several stimuli as LPS, lipoteichoic acid, and glycan peptides, the expression of several genes modi es in monocytes to ensure an adequate body response to invading pathogens [2]. The expression of reported reference genes in macrophages might change in critical clinical conditions such as sepsis [21] Thereby, to express changes in the interest genes, it is rst necessary to determine which reference gene expression is stable in macrophages of septic patients.
A previous report using LPS stimuli in 6 hours cultured macrophages indicated the HNRNPAB as the most stable reference gene [18]. Two other studies reported ACTB as a reference gene in monocytes from septic patients [11,13]. We reported herein HNRNPAB and ACTB as two of the least stable genes, according to GeNorm, NormFinder, and BestKeeper tools.
The other two reports used 18S as a reference gene to normalize interest gene expressions in monocytes [15,26].
The NormFinder and BestKeeper analysis indicated 18S is the most unstable reference gene among 15.
The use of GAPDH to normalize gene expression was reported in mononuclear cells from septic patients [16] and in an experimental septic model using immortalized monocyte [27]. GAPDH was considered herein as the most stable gene by the NormFinder analysis.
Another study with monocytes from critically ill patients and septic patients used the geometric mean of ve targets genes that had the smallest variation among them as reference genes [12]. As mentioned, the majority of these studies used in vitro LPS-stimulated monocytes, which differ from the condition of cells obtained from septic patients.
A study about reference genes in monocyte stimulated with LPS reported that gene expression normalization with ACTB or GAPDH could lead to imprecise results. In certain conditions, such as LPS stimulation, expressions of both genes are modulated and, therefore, cannot be considered as good normalizers of gene expression in these samples [24].

Conclusions
We identi ed STX5A, PGK1, GAPDH, and HPRT1 as the most stable reference genes for the analysis of gene Availability of data and materials The datasets generated and/or analyzed during the current study are not publicly available, but are available from the corresponding author on reasonable request.

Competing interests
The authors declared that no con ict of interest exists.

Funding
We thank the following nancial support:   Average values (M) of expression stability of the 15 reported reference genes analyzed by using the GeNorm algorithm. This graph displays the output of GeNorm plotting the M value against the reference genes considering all samples from each group. Genes with M≤ 0.5 are considered as optimal reference [19].