Comprehensive selection of reference genes for expression studies in meniscus injury using quantitative real-time PCR

    Comprehensive selection of reference genes for expression studies in meniscus injury using quantitative real-time PCR Mariana Ferreira Leal, Gustavo Gonc¸alves Arliani, Diego Costa Astur, Carlos Eduardo Franciozi, Pedro Debieux, Carlos Vicente Andreoli, Mar´ılia Cardoso Smith, Alberto de Castro Pochini, Benno Ejnisman, Moises Cohen PII: DOI: Reference:

S0378-1119(16)30149-4 doi: 10.1016/j.gene.2016.03.005 GENE 41210

To appear in:

Gene

Received date: Revised date: Accepted date:

26 November 2015 11 February 2016 4 March 2016

Please cite this article as: Leal, Mariana Ferreira, Arliani, Gustavo Gon¸calves, Astur, Diego Costa, Franciozi, Carlos Eduardo, Debieux, Pedro, Andreoli, Carlos Vicente, Smith, Mar´ılia Cardoso, de Castro Pochini, Alberto, Ejnisman, Benno, Cohen, Moises, Comprehensive selection of reference genes for expression studies in meniscus injury using quantitative real-time PCR, Gene (2016), doi: 10.1016/j.gene.2016.03.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Comprehensive selection of reference genes for expression studies in meniscus

PT

injury using quantitative real-time PCR

Mariana Ferreira Leal1,2, Gustavo Gonçalves Arliani1, Diego Costa Astur1, Carlos

SC RI

Eduardo Franciozi1, Pedro Debieux1, Carlos Vicente Andreoli1, Marília Cardoso Smith2, Alberto de Castro Pochini1, Benno Ejnisman1, Moises Cohen1.

1

NU

Departamento de Ortopedia e Traumatologia, Universidade Federal de São Paulo,

04038-032, São Paulo, SP, Brazil

MA

2

Disciplina de Genética, Departamento de Morfologia e Genética, Universidade Federal

ED

de São Paulo, 04023-001, São Paulo, SP, Brazil

PT

Corresponding author: Mariana Ferreira Leal, Ph.D. Address: Departamento de Ortopedia e Traumatologia, Universidade Federal de São Paulo, Rua Borges Lagoa,

CE

783, CEP: 04038-032, São Paulo, SP, Brazil. Tel.: +55-11-55716621; Fax: +55-11-

AC

55716621. Email address: [email protected].

ACCEPTED MANUSCRIPT ABSTRACT The meniscus plays critical roles in the knee function. Meniscal tears can lead to knee

PT

osteoarthritis. Gene expression analysis may be a useful tool for understanding meniscus tears, and reverse-transcription quantitative polymerase chain reaction (RT-

SC RI

qPCR) has become an effective method for such studies. However, this technique requires the use of suitable reference genes for data normalization. We evaluated the suitability of six reference genes (18S, ACTB, B2M, GAPDH, HPRT1 and TBP) using

NU

meniscus samples of (1) 19 patients with isolated meniscal tears, (2) 20 patients with

MA

meniscal tears and combined anterior cruciate ligament injury (ACL), and (3) 11 controls without meniscal tears. The stability of the candidate reference genes was determined using the NormFinder, geNorm, BestKeeper DataAssist and RefFinder

ED

software packages and comparative ΔCt method. Overall, HPRT1 was the best single

PT

reference gene. However, GenEx software demonstrated that two or more reference genes should be used for gene expression normalization, which was confirmed when we

CE

evaluated TGFβR1 expression using several reference gene combinations. HPRT1+TBP

AC

was the most frequently identified pair from the analysis of samples of (1) meniscal tear samples of patients with a concomitant ACL tears, (2) all meniscal tears, and (3) all samples. HPRT1+GAPDH was the most frequently identified pair from the analysis of samples of isolated meniscal tear samples and controls. In the analysis involving only controls, GAPDH+18S was the most frequently identified pair. In the analysis of only isolated meniscal tear samples and in the analysis of meniscal tear samples of patients with concomitant ACL tears and controls, both HPRT1+TBP and HPRT1+GAPDH were identified as suitable pairs. If the gene expression study aims to compare noninjured meniscus, isolated meniscal tears and meniscal tears of patients with ACL tears

ACCEPTED MANUSCRIPT as three independent groups, the trio of HPRT1+TBP+GAPDH is the most suitable combination of reference genes.

PT

Abreviation list

SC RI

ACL: anterior cruciate ligament

RT-qPCR: reverse transcription-quantitative polymerase chain reaction

TLDA: TaqMan Low-Density Array

MA

AAV: adeno-associated virus Crt: relative cycle threshold

PT

CV: coefficient of variance

ED

Ct: cycle threshold SD: standard deviation

NU

MRI: magnetic resonance imaging

RQ: relative quantification

AC

CE

Acc.SD: accumulated standard deviation

Keywords: knee injury, meniscal tears, gene expression, reverse-transcription quantitative polymerase chain reaction, expression normalization, reference genes.

ACCEPTED MANUSCRIPT 1. Introduction Menisci are important components in joint biomechanics with crucial roles in the knee

PT

joint: distributing joint forces, load bearing, and enhancing joint stability [1,2]. Lesion of this structure can cause pain, joint swelling, and osteoarthritis in the long term.

SC RI

Younger people are more likely to have acute lesions due to trauma, whereas older people are more likely to have lesions due to degeneration [3,4,5]. Patients with traumatic meniscal tears commonly present an associated rupture of the anterior cruciate

NU

ligament (ACL) [6]. Once present, meniscus tears are associated with an accelerated

MA

progression of cartilage degeneration in the knee compared with individuals with osteoarthritis but without tears [7,8].

Recent studies have been performed to understand the gene expression alterations that

ED

may have a role in human meniscal tears. In a transcriptome analysis, several genes

PT

were identified that were differentially expressed with age and chondrosis in patients with meniscus tears [5]. Moreover, Brophy et al. investigated whether the expression of

CE

osteoarthritis markers (matrix components, cytokines, chemokines, aggrecanases,

AC

metalloproteinases, and transcription factor genes) are age- and sex-related in meniscal tears with and without a concomitant ACL tear [9]. The authors demonstrated that the meniscus in younger patients reacts with an intrinsic response and is more prone to inflammatory changes. Conversely, there were no differences in inflammatory cytokines or chemokines in the group of patients over forty years old [9]. If confirmed in larger studies, these markers may monitor local events at the surgical sites and detect osteoarthritis progression [10]. Investigation of gene expression in human meniscal samples may help in improving the understanding of meniscal tears as well as osteoarthritis progression. Moreover, gene

ACCEPTED MANUSCRIPT expression analysis will be important for guiding patient management and the development of new therapeutic options for these knee afflictions. powerful

techniques,

including

microarrays

and

high-throughput

PT

Although

measurements, have been developed to detect gene expression levels, the reverse

SC RI

transcription-quantitative polymerase chain reaction (RT-qPCR) is commonly used in many laboratories [11]. Moreover, because of its accuracy, sensitivity, and capacity for high-throughput analysis, RT-qPCR is currently considered to be the gold standard

NU

technique for evaluation of gene expression [12]. RT-qPCR is one of the most

MA

commonly utilized approaches in functional genomics research, and its use in gene expression analysis may become more routine; furthermore, this technique is commonly

transcriptomic analysis.

ED

used to validate data obtained by other methods [13], including the data of

PT

To obtain reliable data using RT-qPCR, a common method is to normalize the target gene expression using an endogenous reference gene. Ideally, reference genes should be

CE

stably expressed or at least vary only slightly in expression in all tissues or cells under

AC

the experiment conditions [14]; therefore, a validation experiment for the evaluation of reference gene expression stability for each target tissue and disease is recommended [15,16]. Normalization with unstable internal controls may result in different values, leading to erroneous results [17]. However, many authors do not critically evaluate their RT-qPCR experiments; therefore, the experiments are improperly designed and difficult to repeat because of insufficient data quality [18]. Consequently, the use of suitable reference genes with stable expression in the studied tissue (normal and/or injured) is essential for effective data normalization and the acquisition of accurate and meaningful biological data.

ACCEPTED MANUSCRIPT Suitability of reference genes has been evaluated in some human musculoskeletal diseases, such as shoulder instability [19], rotator cuff tears [20], ACL tears [21],

PT

osteoarthritic articular cartilage (hip and knee) [22], human lumbar vertebral endplate with modic changes [23], and skeletal muscle with chronic degenerative changes [17].

SC RI

To our knowledge, no previous studies have described the best individual or set of reference genes for gene expression analysis in human meniscus samples. A previous study used GAPDH for gene expression normalization in meniscal tear samples of

NU

patients with and without a concomitant ACL tear [9].

MA

In this study, we assessed the suitability of six reference genes frequently reported in the literature (18S, ACTB, B2M, GAPDH, HPRT1 and TBP) using meniscus injured samples of patient with or without concomitant ACL tears as well as meniscus non-

ED

injured samples by analyzing gene stability with five software packages and

PT

comparative delta cycle threshold (ΔCt) method.

AC

2.1. Patients

CE

2. Materials and methods

Tissue samples were obtained from 39 patients with medial meniscal tears, including 19 samples of patients with isolated medial meniscal tears and 20 samples of patients with meniscus injury and a concomitant ACL injury. The following inclusion criteria were employed: age between 18 and 50 years old, clinical history compatible to meniscal injury (such as pain, swelling, stiffness, catching and locking), at least one specific physical examination test positive among McMurray [24], Appley [25]and Steiman [26] tests that were used to diagnose meniscus injury [27], magnetic resonance imaging (MRI) diagnosis of medial meniscus injury with abnormal signal extending to at least one articular surface involving the posterior horn and or the body of the medial meniscus [28],

ACCEPTED MANUSCRIPT and arthroscopic confirmation of the medial meniscus lesion involving its posterior horn and or its body. The Lachman test [29], Anterior Drawer test [30], and Pivot-Shift tests

PT

[31] were used to diagnose ACL injury [32,33]. Coronal and sagital MRI view were used to identify meniscal and ACL lesions. All injuries were confirmed during arthroscopic

SC RI

procedure and reclassified if necessary.

The following exclusion criteria were also applied: medial meniscus lesions treated by suture (outside-in, inside-out or all-inside), medial meniscus stable lesions tested by probe

NU

palpation such as some longitudinal lesion <1cm, radial lesions <5mm, partial thickness

MA

lesions. The stable, unstable criteria was defined intra-operatively by the surgeon. Additionally, 11 patients without any history of meniscal tears were included in this study as a control group. These patients were arthroscopically operated for other knee

ED

injuries, such as isolated ACL injury. All control patients were physically active. Table

PT

1 displays the main clinical outcomes of the studied cases and controls. This study was performed with the approval of the Ethics Committee of the

CE

Universidade Federal de São Paulo (UNIFESP), Brazil (CEP #51436). Written informed

AC

consent with approval of the ethics committee was obtained from all patients prior to specimen collection.

2.2. Tissue samples To collect tissue samples, patients were prepared in the standard fashion for arthroscopy meniscus surgery. A standard arthroscopic joint evaluation was carried out, confirming the meniscus injury or meniscus and ACL injuries diagnosis. During surgery, about 5 mm3 samples of the innermost part of the injured area of the posterior horn and the body of the medial meniscus were collected for gene expression analysis.

ACCEPTED MANUSCRIPT In the controls, a sample fragment of about 5 mm3 was resected from the innermost part of the healthy medial meniscus body by arthroscopy.

PT

All tissue specimens were immediately immersed in Allprotect Tissue Reagent (Qiagen,

SC RI

USA) and stored at -20 °C until RNA extraction.

2.3. RNA extraction

Total RNA was extracted from 10-20 mg of tissue sample using an AllPrep

NU

DNA/RNA/miRNA Mini Kit (Qiagen, USA) according to the manufacturer’s protocol.

MA

The mechanical lysis step was performed using the Tissue Lyser LT equipment (Qiagen, USA). RNA concentration and quality were immediately determined using a Nanodrop ND-1000 (Thermo Scientifc, USA) and the integrity of the RNA was verified by gel

ED

electrophoresis on a 1% agarose gel. Aliquots of the total RNA were stored at -80 °C until

CE

2.4. RT-qPCR

PT

further use.

AC

RT-qPCR gene expression quantifications were performed according to MIQE guidelines [34]. Only RNA samples with the optical density (OD)260/280 > 1.8 were used, following the MIQE protocol. First, cDNA was synthesized from 200 ng of RNA using a High-Capacity cDNA Reverse Transcription Kit (Life Technologies, USA) according to the manufacturer’s protocol. To detect the range of expression of the six candidate reference genes, reactions were performed with 75 ng of cDNA input using TaqMan Low-Density Array (TLDA) cards (Life Technologies, USA) and ViiA 7 Real-Time PCR System (Life Technologies, USA). Only inventoried TaqMan Gene Expression Assays (Life Technologies, USA) were

ACCEPTED MANUSCRIPT chosen for gene expression analysis. The final volume in each TLDA well is approximately 1 μl. All reactions were performed in triplicate.

PT

To identify the best combination of reference genes, we also quantified the mRNA expression of target gene, TGFβR1, using the candidate reference genes for

SC RI

normalization. TGFβRI is considered one of the main receptor of TGFβ and has a key role in the canonical TGFβ signaling pathway [35]. Overexpression of TGFβ via adenoassociated virus (AAV) is capable of modulating the reparative activities of human

NU

meniscal cells, allowing for the healing of meniscal lesions [36]. Therefore, TGFβR1

MA

may also have a role in meniscal tears and healing. Other 9 target genes (TGFβ1, GDF5, COMP, TNC¸ TNXB, FN1, LOX, PLOD1 and PLOD2) were also evaluated to identify the best combination of reference genes (data not shown).

ED

For each sample, the candidate reference and target genes were assayed on the same card

PT

to exclude technical variations. The 6 reference genes and target genes are summarized in Table 2.

CE

The relative threshold method (Crt method) was applied, which is a robust method that

AC

sets a threshold for each curve individually based on the shape of the amplification curve, regardless of the height or variability of the curve during its early baseline fluorescence. The expression of TGFβR1 gene across the samples was calculated using the equation ΔCrt, in which [ΔCrt = target gene (TGFβR1) Crt – the mean of reference genes Crt]. A lower cycle threshold value (Crt) indicates higher gene expression.

2.5. Analysis of reference gene expression stability We categorized the tissue samples into the following 7 groups: 1) isolated meniscal tear samples (N = 19); 2) meniscal tear samples of patients with a concomitant ACL tear (N = 20); 3) meniscal control samples (N = 11); 4) all injured meniscus (N = 39); 5) isolated

ACCEPTED MANUSCRIPT meniscal tear samples and controls (N = 30); 6) meniscal tear samples of patients with a concomitant ACL tear and controls (N = 31); 7) all meniscus samples (N = 50).

PT

Typically, gene expression studies compare transcript levels between case (i.e., the injured tissue) and control samples, therefore we created the groups #5, #6 and #7.

SC RI

However, some researchers have been investigated a possible association between gene expression and clinical variables [9], therefore we created the groups #1, #2 and #4. In addition, the group composed by only controls (group #3) was created since the

NU

understanding of gene expression regulation in non-injured ligaments is still necessary.

MA

For comparisons of candidate reference gene stability we used the software programs NormFinder

(http://www.mdl.dk/publicationsnormfinder.htm),

geNorm

(http://medgen.ugent.be/~jvdesomp/genorm/http://medgen.ugent.be/~jvdesomp/genorm/), (http://www.gene-quantifcation.de/bestkeeper.html)

ED

BestKeeper

and

DataAssist

PT

(http://www.lifetechnologies.com/us/en/home/technical-resources/softwaredownloads/dataassist-software.html) and the comparative ΔCt method [37]. We also used

CE

the RefFinder software (http://www.leonxie.com/referencegene.php) which integrates the

AC

results of geNorm, Normfinder, BestKeeper, and the comparative ΔCt method to compare and rank the tested candidate reference genes. NormFinder accounts for both intra- and inter-group variations when evaluating the stability of each single reference gene [38]. The stability values and standard errors are calculated according to the transcription variation of the reference genes. Stably expressed genes, which have low variation in expression levels, present low stability values. NormFinder analysis also calculated the stability value for two reference genes. geNorm calculates the expression stability value (M) for each gene based on the average pairwise expression ratio between a particular gene and all other reference genes. geNorm sequentially eliminates the gene that shows the highest variation relative to all the other

ACCEPTED MANUSCRIPT genes based on paired expression values in all the studied samples. The most stably expressed gene yields the lowest M value, and then the two most stable reference genes

PT

are determined by stepwise exclusion of the least stable gene [39]. Because of the elimination process, geNorm cannot identify a single suitable reference gene, and ends up

SC RI

by suggesting a pair of genes that shows high correlation and should be suitable for normalization of qPCR studies.

Bestkeeper was used to rank the 6 reference genes based on the standard deviation (SD)

NU

and coefficient of variance (CV) expressed as a percentage of the cycle threshold (Ct)

MA

level [40]. The more stable reference gene presents the lowest CV and SD. Bestkeeper also uses a statistical algorithm wherein the Pearson correlation coefficient for each

significance of the pair.

ED

candidate reference gene pair is calculated along with the probability of correlation

PT

DataAssist software provided a metric to measure reference gene stability based on the geNorm algorithm. In contrast to the other programs, DataAssist uses the relative

CE

quantification (RQ) to calculate the stability value of individual candidate reference

AC

genes. The lower score represents the more stable the control. The comparative ΔCt method is based on the comparing relative expression of pairs of possible reference genes within each sample. The stability of the candidate housekeeping genes is ranked according to reproducibility of the gene expression differences among studied samples. Lastly, RefFinder assigns an appropriate weight to an individual gene and calculated the geometric mean of their weights for the overall final ranking based on the rankings from geNorm, Normfinder, BestKeeper, and the comparative ΔCt. GenEx software (http://genex.gene-quantifcation.info/) was used to determine the optimal number of reference genes by calculating the accumulated standard deviation (Acc.SD). If

ACCEPTED MANUSCRIPT larger number of reference genes is used, random variation among the genes’ expression partially cancel reducing the SD. A minimum in the Acc.SD plot indicate the number of

PT

reference genes that give the lowest SD.

SC RI

2.6. Statistical analysis

To compare TGFβR1 expression between the groups, we first verified the distribution of the data using the Shapiro-Wilk normality test for the determination of the appropriate

NU

tests for the subsequent statistical comparisons. TGFβR1 expression (ΔCrt) was not normally distributed. Therefore, the Mann-Whitney test was performed to compare

MA

TGFβR1 expression between the studied groups. A p-value of < 0.05 was considered

ED

statistically significant.

PT

3. Results

3.1. Reference gene expression levels

CE

The distribution of Crt values for each of the 6 candidate reference genes is shown in

AC

Fig. 1. These genes displayed a wide range of expression levels. 18S presented the highest expression level among the candidate reference genes (mean Crt value ± SD: 12.52 ± 2.34). In contrast, TPB (31.16 ± 1.61) and HPRT1 (31.85 ± 1.79) presented the lowest expression level in meniscal samples.

3.2. Reference gene expression stability Table S1 displays the stability value ranking of single candidate reference genes as determined by the different software packages and comparative ΔCt method. In our analysis, all the reference genes for all analysis groups presented M values less than the geNorm threshold of 1.5, which is recognized as stable (Table S1). However, 18S,

ACCEPTED MANUSCRIPT ACTB, and B2M presented high SD (±x-fold) of Crt in the analysis involving all samples according to BestKeeper software [SD (±x-fold) = 3.21, 3.64, and 3.41,

PT

respectively), whereby any studied gene with SD (±x-fold) >3 can be considered inconsistent. Furthermore, 18S, ACTB, and B2M presented high SD (±x-fold) in the

SC RI

analysis of (1) all meniscal tear samples, (2) isolated meniscal tears, and (3) isolated meniscal tears and controls. Moreover, ACTB and B2M presented high SD (±x-fold) in the analysis of (1) meniscal tear samples of patients with concomitant ACL tears and (2)

NU

meniscal tear samples of patients with concomitant ACL tears and controls. ACTB demonstrated high SD (±x-fold) in the analysis involving only controls.

MA

Although none of the software packages and comparative ΔCt method suggested the same rank of reference genes in the studied sample groups, the methods applied did

ED

generate similar rankings of reference gene stability for each analysis group (Table S1).

PT

Table 3 shows the best suitable reference gene by the different methods applied. In the present study, HPRT1 was the most suitable reference gene for the meniscus samples.

CE

As previously described, gene expression studies typically compare transcript levels

AC

between injured and non-injured tissue samples. When the isolated meniscus tear samples and controls were evaluated together, HPRT1 and GAPDH were the most suitable reference genes. When considering the meniscal tear samples of patients with a concomitant ACL tear and controls, TBP followed by GAPDH was the most stable gene. When considering all injured meniscal tear samples and controls (all samples), HPRT1 followed by TBP was also the most stable gene (Table 3; Table S1). When we individually evaluated each group of meniscus samples, we observed that HPRT1 followed by TBP was the most stable gene for the isolated meniscal tear samples as well as for meniscal tear samples of patients with a concomitant ACL tear. In the control group, GAPDH and 18S were the most stable genes. HPRT1 followed by

ACCEPTED MANUSCRIPT TBP was the most suitable gene in the analysis involving all meniscal tear samples

3.3. Analysis of the best combinations of reference genes

PT

(Table 3; Table S1).

SC RI

Table 4 displays the best combinations of reference genes, as suggested by the software packages, comparative ΔCt method, and by visual inspection of all the ranks generated by these analyses. Overall, HPRT1 + TBP and HPRT1 + GAPDH pairs of genes were

NU

the most frequently identified pairs. HPRT1 + TBP was the most frequently identified

MA

pair from the analysis of samples from (1) meniscal tear samples of patients with a concomitant ACL tear, (2) all meniscal tears, and (3) all samples. HPRT1 + GAPDH was the most frequently identified pair from the analysis of samples from isolated

ED

meniscal tear samples and controls. In the analysis involving only controls, GAPDH +

PT

18S was the most frequently identified pair. In the analysis of only isolated meniscal tear samples and in the analysis of meniscal tear samples of patients with concomitant

AC

pair (Table 4).

CE

ACL tears and controls, no more than two methods agreed in the definition of the best

The NormFinder, GeNorm, DataAssist, and BestKeeper software packages only indicated up to 2 genes as the best combination of reference genes. By visual inspection of all the ranks generated, software used, and comparative ΔCt method, we observed that HPRT1 + TBP + GAPDH was more frequently the best trio of reference genes. HPRT1 + TBP + 18S was the best trio only in the analysis of meniscal tear samples from patients with concomitant ACL tears. Additionally, GAPDH + TBP + 18S was the best trio in the analysis involving only control samples. We used the GenEx software package to determine the appropriate number of reference genes for a reliable normalization. In this analysis, the optimal number of reference

ACCEPTED MANUSCRIPT genes is indicated by the lowest SD. In all analyses, the Acc.SD of two reference genes did not differ by >0.1 from the observed metric when using more than two genes (Fig.

PT

2). However, one reference gene is not suitable for gene expression normalization in the analysis involving (1) isolated meniscal tears and controls or (2) meniscal tears of

SC RI

patients with concomitant ACL tears and controls. In these groups of samples, the Acc.SD of one reference gene was >0.1 from the observed metric when using two or more genes (Fig. 2). Moreover, in the analysis involving all meniscal samples, the

NU

Acc.SD of one reference gene was >0.1 from the observed metric when using four or

MA

more genes (Fig. 2). In the analysis of meniscal tears of patients with concomitant ACL tears, the Acc.SD of one reference gene was also >0.1 from that observed when using

ED

more than three genes, which presented the lowest Acc. SD.

PT

3.4 Effects of reference gene choice

To evaluate the effect of appropriate reference gene selection, an expression analysis

CE

was performed by comparing data from 1) meniscal tears from patients with and

AC

without a concomitant ACL tear, 2) isolated meniscal tear samples and controls, 3) meniscal tear samples of patients with a concomitant ACL tear and controls, and 4) injured meniscus samples and controls. This analysis was performed using TGFβR1 as the target gene in all the analyses. For reference genes, we used the most frequently identified pairs described above (HPRT1 + TBP, GAPDH + 18S, HPRT1 + GAPDH, and GAPDH + TBP). We also performed collagen genes expression analysis using three reference genes (HPRT1 + TBP + GAPDH, HPRT1 + TBP + 18S, and GAPDH + TBP + 18S), four reference genes (HPRT1 + TBP + GAPDH + 18S), and all studied reference genes. TGFβR1 was also normalized by only 18S, ACTB, GAPDH, and HPRT1, as commonly described in the literature involving joint lesions. B2M + 18S

ACCEPTED MANUSCRIPT was used as an example of a not suitable reference gene pair for gene expression normalization in meniscus samples.

PT

Although the normalized expression quantities differed between the various combinations of reference genes, the distributions of the target gene expression in the

SC RI

studied samples were similar (Fig. 3).

Table 5 shows TGFβR1 expression using the different reference gene combinations for data normalization. TGFβR1 expression was significantly increased in the isolated

NU

meniscal tear samples compared with the controls when using several reference gene

MA

combinations (p < 0.05), except for HPRT1 (p = 0.121), HPRT1 + TBP (p = 0.061), and B2M+18 (p = 0.077).

Alternatively, TGFβR1 expression was significantly increased in the meniscal tear

ED

samples from patients with concomitant ACL tears compared with the controls only

PT

when using GAPDH (p = 0.0212) as a reference gene. Moreover, all injured meniscal samples presented increased TGFβR1 expression compared with controls only when the

CE

target expression was normalized by GAPDH (p = 0.008) or GAPDH + 18S (p = 0.043).

AC

When we compared isolated meniscus tear samples and meniscal samples from patients with ACL tears, TGFβR1 was significantly different between groups when its expression was normalized by B2M + 18S (p = 0.049).

4. Discussion Comparing gene expression in different samples may generate misleading information related to distinctive amounts of material, RNA extraction and efficiency of reverse transcription. To overcome those risk potentials and be accurate, RT-qPCR relies on normalisation to an internal control, often referred to as a housekeeping gene. Housekeeping genes are constitutive genes which are transcribed at a relatively constant

ACCEPTED MANUSCRIPT level. Its products are vital to the basic functions of a cell. It is generally assumed that their expression is minimally affected by experimental conditions. Related to these

PT

characteristics, some candidates housekeeping genes are used on various studies to have its stability ranked according to the reproducibility of differences in gene expression

SC RI

among the studied samples. This stability analysis is helps to select the most appropriate candidate reference gene [19,20,21].

Reference genes have been described for RT-qPCR studies in several diseases and

NU

tissues [17,22,23,41,42,43,44,45], and our group recently identified the most stable

MA

reference genes in the glenohumeral capsule of patients with and without shoulder instability [19], in tendon of patients with and without rotator cuff tears [20] and in ACL of patients with or without ACL tears [21]. To the best of our knowledge, no prior

ED

study has aimed to identify suitable reference genes for gene expression analyses by

PT

quantitative approaches in the human meniscus. In the present study, we used five software packages (NormFinder, geNorm,

CE

BestKeeper, DataAssist, and RefFinder) and the comparative ΔCt method to evaluate

AC

the stability of reference gene expression. As each analysis uses distinct algorithms, different results can be expected. Therefore, it is important to use more than one software package/method to identify the most suitable reference genes among a set of candidates. Although the distinct analysis differed in the rankings of reference gene stability as well as in the identity of the most suitable pair, at least two programs produced results that agreed for almost all the analyses. Our results demonstrate that the use of 5 statistical tools and comparative ΔCt method aids in the identification of the best reference genes. Surprisingly, Normfinder, geNorm, and BestKeeper from RefFinder did not lead to the same output as we got from our NormFinder, geNorm, and Bestkeeper interface (data

ACCEPTED MANUSCRIPT not shown), which is probably due to different versions of the algorithm. This lack of agreement was previously reported in literature [46].

PT

All the reference genes in this study were recognized as stable by geNorm analysis under the different experimental conditions tested. However, 18S, ACTB, and B2M were

SC RI

not suitable by BestKeeper software in most of the groups of analyses. According to BestKeeper analysis, these reference genes should not be used in studies involving isolated meniscal tears, meniscal tears of patients with and without concomitant ACL

NU

injury, as well as in a study involving all types of meniscal samples investigated here.

MA

In the different groups of analyses, HPRT1 appeared to be the most suitable gene overall. GAPDH was also previously used as a reference gene in a gene expression study from human meniscal tears with and without concomitant ACL injury [9];

ED

however, this gene was the most stable only in some analyses with meniscal control

PT

samples and with isolated meniscal tear samples and controls. It is increasingly clear that in most situations, a single reference gene is not sufficiently

CE

stable [47]. Here, we observed that the use of one reference gene is not appropriate,

AC

mainly for the analyses involving meniscal tear samples and controls. In our study, a pair of reference genes seems to be suitable for gene expression normalization in the studied groups. When a larger number of reference genes is used, the SD of the normalization factor (mean of reference gene expression) is reduced, and the random variation among the expression of the tested genes is partially canceled. As the inclusion of additional reference genes increases the time and money required for the analysis, it is important to consider the degree of improvement and overall noise contributed by reference genes when deciding how many reference genes are required. Considering the reproducibility of real-time PCR equipment, we believe that the use of several reference genes does not significantly improve the data quality. However, it is

ACCEPTED MANUSCRIPT important to highlight that we observed the use of one, two, three, or more reference genes may lead to differences in the statistical analysis result of some group

PT

comparisons. Although different combinations of reference genes were determined as being the most

SC RI

suitable for the various analysis, HPRT1 + TBP and HPRT1 + GAPDH were the most frequently identified pair, and HPRT1 + TBP + GAPDH was the most frequently identified trio. The selection of the appropriate pair or trio should consider the group of

NU

meniscus samples that will be investigated. For example, HPRT1 + TBP + 18S was the

MA

best trio only in the analysis of meniscal tear samples of patients with concomitant ACL tears, and GAPDH + TBP + 18S was the best trio in the analysis involving only control samples, which is in agreement with the observation that 18S may be stable in these

ED

groups according to BestKeeper software.

PT

To identify the best combination of reference genes, we evaluated TGFβR1 expression in samples from meniscus tissue of the cases and the controls. Statistical comparison

CE

revealed that TGFβR1 expression differed between the isolated meniscal tear samples

AC

and the controls when using several reference gene combinations (p < 0.05), except for HPRT1, HPRT1 + TBP, and B2M + 18. This finding reinforces the use of one reference gene (even the most stable) is not appropriate for gene expression normalization in this group of samples. Moreover, HPRT1 + TBP and B2M + 18 were not the best pair in this group of samples. Only geNorm suggested that HPRT1 + TBP were the most stable genes in the analysis involving isolated meniscal tear samples and controls. HPRT1 + TBP was the best pair of reference genes in the analysis involving all injured meniscus samples and controls, but not when meniscus tear samples from patients with or without ACL injury were grouped separately with controls. Consequently, if a gene expression study aims to compare non-injured menisci, isolated meniscal tears and

ACCEPTED MANUSCRIPT meniscal tears of patients with ACL tears as three independent groups of analysis, HPRT1 + TBP + GAPDH seems to be the most suitable combination of reference

PT

genes. This trio of reference genes was the most suitable for the different comparisons involving injured and non-injured meniscus samples, as well as in the analysis

SC RI

involving injured meniscus samples of patients with and without ACL tears. When meniscal tear samples of patients with concomitant ACL tears were compared with the controls, no significant difference was detected, except when TGFβR1

NU

expression was normalized only by GAPDH. The use of GAPDH as well as GAPDH +

MA

18S for normalization also lead to the observation that TGFβR1 expression differed between meniscal tear samples and controls. As already discussed, these analyses highlight the necessity of the selection of suitable reference genes based on the studied

ED

meniscus samples.

PT

During this study, we also evaluated the use of different reference gene combinations in the expression of other nine extracellular matrix genes (data not show). No significance

AC

combinations.

CE

difference was detected between the studied groups using the different reference gene

Our study presented some limitations. First, we only included a limited number of candidate reference genes, and it is likely that some other genes may also be used as internal references for gene expression studies in meniscus samples from patients with or without history of a meniscal tear. Second, the number of samples available for the Mann-Whitney test was reduced, especially in the control group. However, to the best of our knowledge, no prior study evaluated RNA expression in human non-injured meniscus samples. Nevertheless, it is important to highlight that our results may be relevant to the study of meniscal tears, as well as to the study of normal menisci.

ACCEPTED MANUSCRIPT 5. Conclusions The use of suitable reference genes for a reliable gene expression evaluation using RT-

PT

qPCR should consider the type of meniscus samples investigated (injured or noninjured). HPRT1 was the most suitable reference gene. However, the use of only one

SC RI

reference gene does not seem suitable for gene expression normalization in meniscal tear studies. HPRT1 + TBP and HPRT1 + TBP + GAPDH were the best combination of reference genes for the analysis of involving meniscus samples. However, if the gene

NU

expression study aims to compare non-injured menisci, isolated meniscal tears and

MA

meniscal tears from patients with ACL tears as three independent groups, HPRT1 + TBP + GAPDH is the most suitable combination of reference genes. The results of this

Acknowledgements

PT

expression quantification.

ED

work may benefit future studies of the meniscus that require more accurate gene

CE

We are grateful to scientific support of Sintia Iole Belangero, Ph. D., and Fernanda

AC

Wisnieski, Ph. D., from the Genetic Division of the Federal University of São Paulo (UNIFESP). We are also grateful to Maria Laura Salgado.

References 1. Lee DW, Ha JK, Kim JG (2014) Medial meniscus posterior root tear: a comprehensive review. Knee Surg Relat Res 26: 125-134. 2. Kaleka CC, Debieux P, da Costa Astur D, Arliani GG, Cohen M (2014) Updates in biological therapies for knee injuries: menisci. Curr Rev Musculoskelet Med 7: 247-255.

ACCEPTED MANUSCRIPT 3. Englund M, Guermazi A, Lohmander SL (2009) The role of the meniscus in knee osteoarthritis: a cause or consequence? Radiol Clin North Am 47: 703-712.

PT

4. Pauli C, Grogan SP, Patil S, Otsuki S, Hasegawa A, et al. (2011) Macroscopic and

Osteoarthritis Cartilage 19: 1132-1141.

SC RI

histopathologic analysis of human knee menisci in aging and osteoarthritis.

5. Rai MF, Patra D, Sandell LJ, Brophy RH (2013) Transcriptome analysis of injured human meniscus reveals a distinct phenotype of meniscus degeneration with

NU

aging. Arthritis Rheum 65: 2090-2101.

MA

6. Poulsen MR, Johnson DL (2011) Meniscal injuries in the young, athletically active patient. Phys Sportsmed 39: 123-130.

7. Biswal S, Hastie T, Andriacchi TP, Bergman GA, Dillingham MF, et al. (2002) Risk

ED

factors for progressive cartilage loss in the knee: a longitudinal magnetic

PT

resonance imaging study in forty-three patients. Arthritis Rheum 46: 2884-2892. 8. Hunter DJ, Zhang YQ, Niu JB, Tu X, Amin S, et al. (2006) The association of

CE

meniscal pathologic changes with cartilage loss in symptomatic knee

AC

osteoarthritis. Arthritis Rheum 54: 795-801. 9. Brophy RH, Rai MF, Zhang Z, Torgomyan A, Sandell LJ (2012) Molecular analysis of age and sex-related gene expression in meniscal tears with and without a concomitant anterior cruciate ligament tear. J Bone Joint Surg Am 94: 385-393. 10. Kambic H (2012) The meniscus--a key to unfolding the mystery of osteoarthritis: commentary on an article by Robert H. Brophy, MD, et al.: "Molecular analysis of age and sex-related gene expression in meniscal tears with and without a concomitant anterior cruciate ligament tear". J Bone Joint Surg Am 94: e31. 11. Wang HL, Li L, Tang S, Yuan C, Tian Q, et al. (2015) Evaluation of Appropriate Reference Genes for Reverse Transcription-Quantitative PCR Studies in

ACCEPTED MANUSCRIPT Different Tissues of a Desert Poplar via Comparision of Different Algorithms. Int J Mol Sci 16: 20468-20491.

PT

12. Derveaux S, Vandesompele J, Hellemans J (2010) How to do successful gene expression analysis using real-time PCR. Methods 50: 227-230.

SC RI

13. Kozera B, Rapacz M (2013) Reference genes in real-time PCR. J Appl Genet 54: 391-406.

14. Li YL, Ye F, Hu Y, Lu WG, Xie X (2009) Identification of suitable reference genes

NU

for gene expression studies of human serous ovarian cancer by real-time

MA

polymerase chain reaction. Anal Biochem 394: 110-116. 15. Bustin SA, Mueller R (2005) Real-time reverse transcription PCR (qRT-PCR) and its potential use in clinical diagnosis. Clin Sci (Lond) 109: 365-379.

ED

16. Hruz T, Wyss M, Docquier M, Pfaffl MW, Masanetz S, et al. (2011) RefGenes:

PT

identification of reliable and condition specific reference genes for RT-qPCR data normalization. BMC Genomics 12: 156.

CE

17. Yuzbasioglu A, Onbasilar I, Kocaefe C, Ozguc M (2010) Assessment of

AC

housekeeping genes for use in normalization of real time PCR in skeletal muscle with chronic degenerative changes. Exp Mol Pathol 88: 326-329. 18. Bustin SA (2010) Why the need for qPCR publication guidelines?--The case for MIQE. Methods 50: 217-226. 19. Leal MF, Belangero PS, Cohen C, Figueiredo EA, Loyola LC, et al. (2014) Identification of suitable reference genes for gene expression studies of shoulder instability. PLoS One 9: e105002. 20. Leal MF, Belangero PS, Figueiredo EA, Cohen C, Loyola LC, et al. (2015) Identification of suitable reference genes for gene expression studies in tendons from patients with rotator cuff tear. PLoS One 10: e0118821.

ACCEPTED MANUSCRIPT 21. Leal MF, Astur DC, Debieux P, Arliani GG, Silveira Franciozi CE, et al. (2015) Identification of Suitable Reference Genes for Investigating Gene Expression in

PT

Anterior Cruciate Ligament Injury by Using Reverse Transcription-Quantitative PCR. PLoS One 10: e0133323.

SC RI

22. Pombo-Suarez M, Calaza M, Gomez-Reino JJ, Gonzalez A (2008) Reference genes for normalization of gene expression studies in human osteoarthritic articular cartilage. BMC Mol Biol 9: 17.

NU

23. Zhou ZJ, Zhang JF, Xia P, Wang JY, Chen S, et al. (2014) Selection of suitable

MA

reference genes for normalization of quantitative real-time polymerase chain reaction in human cartilage endplate of the lumbar spine. PLoS One 9: e88892. 24. McMurray TP (1949) The semilunar cartilages. Br J Surg 29: 1.

ED

25. Apley AG (1947) The diagnosis of meniscal injuries: Some new clinical methods. J

PT

Bone Joint Surg 29: 7.

26. Tria AJ, Klein KS (1992) An illustrated guide to the knee. New York: Churchill

CE

Livingstone.

AC

27. Speziali A, Placella G, Tei MM, Georgoulis A, Cerulli G (2015) Diagnostic value of the clinical investigation in acute meniscal tears combined with anterior cruciate ligament injury using arthroscopic findings as golden standard. Musculoskelet Surg. 28. Crues JV, 3rd, Mink J, Levy TL, Lotysch M, Stoller DW (1987) Meniscal tears of the knee: accuracy of MR imaging. Radiology 164: 445-448. 29. Torg JS, Conrad W, Kalen V (1976) Clinical diagnosis of anterior cruciate ligament instability in the athlete. Am J Sports Med 4: 84-93. 30. Marshall JL, Wang JB, Furman W, Girgis FG, Warren R (1975) The anterior drawer sign: what is it? J Sports Med 3: 152-158.

ACCEPTED MANUSCRIPT 31. Galway R, Beaupre A, McIntosh DL (1972) Pivot Shift – a clinical sign of symptomatic anterior cruciate insufficiency. J Bone Joint Surg (Br) 54: 2.

PT

32. Astur DC, Lauxen D, Ejnisman B, Cohen M (2014) Twin athlete brothers with open physes operated for ACL reconstruction on the same day, but with different

SC RI

elapsed times after injury: a 5-year follow-up. BMJ Case Rep 2014. 33. Astur DC, Aleluia V, Veronese C, Astur N, Oliveira SG, et al. (2014) A prospective double blinded randomized study of anterior cruciate ligament reconstruction

NU

with hamstrings tendon and spinal anesthesia with or without femoral nerve

MA

block. Knee 21: 911-915.

34. Taylor SC, Mrkusich EM (2014) The state of RT-quantitative PCR: firsthand observations of implementation of minimum information for the publication of

ED

quantitative real-time PCR experiments (MIQE). J Mol Microbiol Biotechnol

PT

24: 46-52.

35. Moore-Smith L, Pasche B TGFBR1 signaling and breast cancer. J Mammary Gland

CE

Biol Neoplasia 16: 89-95.

AC

36. Cucchiarini M, Schmidt K, Frisch J, Kohn D, Madry H (2015) Overexpression of TGF-beta via rAAV-Mediated Gene Transfer Promotes the Healing of Human Meniscal Lesions Ex Vivo on Explanted Menisci. Am J Sports Med. 37. Silver N, Best S, Jiang J, Thein SL (2006) Selection of housekeeping genes for gene expression studies in human reticulocytes using real-time PCR. BMC Mol Biol 7: 33. 38. Andersen CL, Jensen JL, Orntoft TF (2004) Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Res 64: 5245-5250.

ACCEPTED MANUSCRIPT 39. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, et al. (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric

PT

averaging of multiple internal control genes. Genome Biol 3: RESEARCH0034. 40. Pfaffl MW, Tichopad A, Prgomet C, Neuvians TP (2004) Determination of stable

SC RI

housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper--Excel-based tool using pair-wise correlations. Biotechnol Lett 26: 509-515.

NU

41. Lyng MB, Laenkholm AV, Pallisgaard N, Ditzel HJ (2008) Identification of genes

MA

for normalization of real-time RT-PCR data in breast carcinomas. BMC Cancer 8: 20.

42. Rubie C, Kempf K, Hans J, Su T, Tilton B, et al. (2005) Housekeeping gene

ED

variability in normal and cancerous colorectal, pancreatic, esophageal, gastric

PT

and hepatic tissues. Mol Cell Probes 19: 101-109. 43. Fu J, Bian L, Zhao L, Dong Z, Gao X, et al. (2010) Identification of genes for

CE

normalization of quantitative real-time PCR data in ovarian tissues. Acta

AC

Biochim Biophys Sin (Shanghai) 42: 568-574. 44. Wang Q, Ishikawa T, Michiue T, Zhu BL, Guan DW, et al. (2012) Stability of endogenous reference genes in postmortem human brains for normalization of quantitative real-time PCR data: comprehensive evaluation using geNorm, NormFinder, and BestKeeper. Int J Legal Med 126: 943-952. 45. Wisnieski F, Calcagno DQ, Leal MF, dos Santos LC, Gigek Cde O, et al. (2013) Reference genes for quantitative RT-PCR data in gastric tissues and cell lines. World J Gastroenterol 19: 7121-7128. 46. Llanos A, François JM, Parrou J-L Tracking the best reference genes for RT-qPCR data normalization in filamentous fungi. BMC Genomics 16.

ACCEPTED MANUSCRIPT 47. de Jonge HJ, Fehrmann RS, de Bont ES, Hofstra RM, Gerbens F, et al. (2007)

AC

CE

PT

ED

MA

NU

SC RI

PT

Evidence based selection of housekeeping genes. PLoS One 2: e898.

ACCEPTED MANUSCRIPT Figures Legends Fig. 1. RT-qPCR detection of the expression levels of six reference genes. A lower

PT

cycle threshold value (Crt) indicates higher gene expression. Horizontal line: mean Crt value. MI: isolated meniscal tear samples; ML: meniscal tear samples of patients with

SC RI

concomitant anterior cruciate ligament tear; MC: meniscus samples of controls.

Fig. 2. Accumulated standard deviation for the 6 reference genes in meniscus samples.

NU

Lower values of accumulated standard deviation (Acc.SD) indicate the optimal number

MA

of reference gene as estimated by the GenEx software package. MI: isolated meniscal tear samples; ML: meniscal tear samples of patients with concomitant anterior cruciate

ED

ligament tear; MC: meniscus samples of controls.

PT

Fig. 3. TGFβR1 expression normalized by different combinations of candidate reference genes in meniscus specimens. Data is show as median and interquartile range (error

CE

bars). 18: TGFβR1 expression normalized by 18S; A: TGFβR1 expression normalized

AC

by ACTB; G: TGFβR1 expression normalized by GAPDH; H: TGFβR1 expression normalized by HPRT1; H_T: TGFβR1 expression normalized by HPRT1 + TBP; G_18: TGFβR1 expression normalized by GAPDH + 18S; H_G: TGFβR1 expression normalized by HPRT1 + GAPDH; G_T: TGFβR1 expression normalized by GAPDH + TBP; H_T_G: TGFβR1 expression normalized by HPRT1 + TBP + GAPDH; H_T_18: TGFβR1 expression normalized by HPRT1 + TBP + 18S; G_T_18: TGFβR1 expression normalized by GAPDH + TBP + 18S; H_T_G_18: TGFβR1 expression normalized by HPRT1 + TBP + GAPDH + 18S; ALL: TGFβR1 expression normalized by HPRT1 + TBP + GAPDH + 18S + B2M + ACTB; B_18: TGFβR1 expression normalized by B2M

ACCEPTED MANUSCRIPT + 18S. MI: isolated meniscal tear samples; ML: meniscal tear samples of patients with

AC

CE

PT

ED

MA

NU

SC RI

PT

concomitant anterior cruciate ligament tear; MC: meniscus samples of controls.

AC

Figure 1

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

AC

Figure 2

CE

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

CE

AC

Figure 3

PT

ED

MA

NU

SC RI

PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT Table 1. Distribution of the clinical outcomes of meniscal tear patients and controls. Variable

Cases (N = 39)

Controls (N = 11)

39.5 ± 12.6

29.4 ± 7.3

79.5%

81.8%

PT

Age at surgery, years (mean ± SD)

Duration of condition, months (mean ± SD) Mechanism (% of traumatic onset of symptoms)

AC

CE

PT

ED

MA

NU

N: number of samples; SD: standard deviation.

SC RI

Gender (% of male)

9.4 ± 10.7

4.5 ± 3.6

84.6%

90.9%

ACCEPTED MANUSCRIPT Table 2. Summary of six reference genes and target genes. Gene Name

Gene function

18S

Eukaryotic 18S rRNA

Ribosome subunit

ACTB

Beta-actin

Assay*

PT

symbol Hs99999901_s1

SC RI

Cytoskeletal structural

Hs01060665_g1

protein

Beta-chain of major

B2M

Beta-2-microglobulin

NU

histocompatibility

Hs00984230_m1

MA

complex class I molecules

Oxidoreductase in

ED

Glyceraldehyde-3-phosphate GAPDH

glycolysis and

Hs02758991_g1

PT

dehydrogenase

Hypoxanthine HPRT1

Purine synthesis in Hs02800695_m1 salvage pathway RNA polymerase II,

TATA box binding protein

AC

TBP

CE

phosphoribosyl-transferase

gluconeogenesis

Hs00427620_m1 transcription factor

Transforming growth factor,

Extracellular matrix

beta receptor 1

structural protein

TGFβR1

Hs00610320_m1

*TaqMan probes were purchased as assays-on-demand products for gene expression (Life Technologies, USA).

ACCEPTED MANUSCRIPT Table 3. Best reference gene for each group of sample. Groups

Best reference gene by each method GeNorm

BestKeeper

HPRT1

HPRT1

TBP/ACTB

HPRT1

HPRT1

HPRT1

HPRT1/TBP

GAPDH

18S

HPRT1

Isolated meniscal

HPRT1

HPRT1

TBP

HPRT1

HPRT1

GAPDH/18S

HPRT1

18S

TBP

HPRT1/TBP

HPRT1

HPRT1

HPRT1

GAPDH*

HPRT1/TBP

HPRT1

HPRT1

GAPDH

TBP

HPRT1/TBP

TBP

18S

GAPDH

HPRT1

HPRT1/TBP

HPRT1

HPRT1

HPRT1

SC RI

tear samples Meniscal tear samples of

NU

patients with a concomitant ACL

samples All meniscal tear

Isolated meniscal

GAPDH

GAPDH

AC

Meniscal tear

CE

tear samples and controls

HPRT1

PT

samples

MA

tear Meniscal control

RefFinder

ED

method

DataAssist

PT

NormFindera

ΔCt

samples of

patients with a

concomitant ACL tear and controls All meniscal

HPRT1

samples a

Best combination of two genes determined considering the intragroup and intergroup

variation. *When the intragroup and intergroup variation was not considered, HPRT1 was the best reference gene by NormFinder. Bold letters: best pairs of reference genes

ACCEPTED MANUSCRIPT by more than one of the methods commonly used (ΔCt method, Normfinder, GeNorm,

AC

CE

PT

ED

MA

NU

SC RI

PT

BestKeeper, DataAssist and RefFinder). ACL: anterior cruciate ligament

ACCEPTED MANUSCRIPT Table 4. Best combination of reference genes for each group of sample. Best pair of reference genes by software NormFinde r

GeNorm

a

BestKee per

DataAssi

b

Top genes

Top genes

Best pair of

Best trio of

by ΔCt

by

reference

reference

st

method

RefFinder

genes

c

PT

Groups

genesc

Isolated

HPRT1+G

ACTB+T

B2M+18

ACTB+T

HPRT1+G

HPRT1+TB

HPRT1+TB

HPRT1+TBP+

menisca

APDH

BP

S

BP

APDH

P

P

GAPDH

HPRT1+TB

HPRT1+

ACTB+T

HPRT1+

HPRT1+TB

HPRT1+TB

HPRT1+TB

HPRT1+TBP+1

P

TBP

BP

TBP

P

P

P

8S

Menisca

GAPDH+T

GAPDH

l control

BP

+18S

SC RI

l tear samples Menisca l tear

NU

samples of

MA

patients with a concomi

ACL

l tear samples Isolated menisca

GAPDH+T

GAPDH+T

GAPDH+1

GAPDH+TBP+

+18S

+18S

BP

BP

8S

18S

HPRT1+18

HPRT1+

ACTB+T

ACTB+T

HPRT1+TB

HPRT1+TB

HPRT1+TB

HPRT1+TBP+

S*

TBP

BP

BP

P

P

P

GAPDH

GAPDH+T

HPRT1+

B2M+18

ACTB+T

HPRT1+G

HPRT1+G

HPRT1+G

HPRT1+TBP+

BP**

TBP

S

BP

APDH

APDH

APDH

GAPDH

TBP+18S

HPRT1+

HPRT1+

GAPDH

HPRT1+G

HPRT1+G

TBP+GAP

HPRT1+TBP+

TBP

TBP

+18S

APDH

APDH

DH

GAPDH

AC

menisca

CE

samples All

GAPDH

GAPDH

PT

tear

ED

tant

l tear samples and controls Menisca l tear samples of patients with a

ACCEPTED MANUSCRIPT concomi tant ACL

controls

menisca

HPRT1+TB

HPRT1+

ACTB+T

ACTB+T

HPRT1+G

HPRT1+G

HPRT1+TB

HPRT1+TBP+

P

TBP

BP

BP

APDH

APDH

P

GAPDH

SC RI

All

l samples a

PT

tear and

Best combination of two genes determined considering the intragroup and intergroup

NU

variation; bBest combination of two genes determined considering the correlation values

MA

(r); cBest combination is based in a visual inspection of all the ranks generated by the four software. *When the intragroup and intergroup variation was not considered, HPRT1+TBP was the combination of reference genes by NormFinder. ** When the

ED

intragroup and intergroup variation was not considered, HPRT1+GAPDH was the

PT

combination of reference genes by NormFinder. Bold letters: best pairs of reference genes by more than one of the methods commonly used (Normfinder, GeNorm,

CE

BestKeeper, DataAssist, ΔCt method and RefFinder). Underlined letters: best pairs of

AC

reference genes by visual inspection and of the methods commonly used. ACL: anterior cruciate ligament.

ACCEPTED MANUSCRIPT Table 5. TGFβR1 expression normalized by different combinations of reference genes in the meniscus samples. TGFβR1 expression [ΔCrt: median (IQR)]a MI

ML

p-

MI

MC

value 16.5

8

0.054

16.49

0.045

7

8

(0.39)

*

(1.02

(0.85

(1.02

)

)

)

4.92

5.62

(1.62

16.5

16.49

p-

MI +

7

value

ML

1

16.3

16.49

4

(0.39)

(0.39)

(0.85

(0.96

)

)

(1.62

(0.72)

0

)

0

4.52

4.91

5.22

0.013

4.91

5.22

0.022

4.77

5.22

0.008

(1.03

(0.63

(1.03

(0.87)

*

(0.63

(0.87)

*

(0.76

(0.87)

*

)

)

)

-2.83

-2.49

-2.75

-2.53

0.303

(1.32

(1.13

)

)

HPRT1

-3.21

-2.71

+ TBP

(1.22

(1.51

)

)

10.3

+ 18S

2

10.7

MA

-2.83

)

-2.53

0.121

ED 0.261

0.095

5

AC

GAPDH

0.448

5.87

(2.07

(0.72)

)

)

-2.49

-2.53

0.741

(1.15

(1.13

(1.15

(1.29

(1.15

)

2)

)

2)

)

2)

-2.71

-2.75

-3.08

-2.75

(1.51

(0.87)

(1.39

(0.87)

-3.21

-2.75

(1.22

(0.87)

0.061

)

0.563

)

10.3

10.95

0.017

10.7

10.95

2

(0.49)

*

5

(0.49)

0.201

10.5

10.95

0.043

7

(0.49)

*

1.02

1.27

0.066

(0.83

(0.84)

(0.39

(0.79

(0.39

(0.82

)

)

)

)

)

0.83

1.16

+

(0.92

GAPDH

0.144

0.83

1.27

0.024

1.16

1.27

(0.72

(0.92

(0.84)

*

(0.72

(0.84)

)

)

)

GAPDH

0.49

1.04

+ TBP

(1.07

0.286

0.283

)

)

0.49

1.19

0.045

1.04

1.19

(0.86

(1.07

(0.69)

*

(0.86

(0.69)

)

)

)

HPRT1

-0.67

-0.10

+ TBP

(1.12

+

0.249

0.127

)

-0.67

-0.05

(0.97

(1.12

(0.82)

)

)

)

3.19

3.64

0.045

0.178

)

(0.79

HPRT1

0.331

)

(1.32

PT

HPRT1

4.52

(0.72)

5.19

0.276

(2.16

0.238

0.773

value

0.013

(2.16

5.87

p-

5.87

*

5.62

MC

4.92

GAPDH

0.613

MC

value 16.0

CE

ACTB

16.0

ML

NU

18S

p-

SC RI

ce genes

PT

Referen

0.69

1.19

(1.04

(0.69)

0.051

)

-0.10

-0.05

(0.97

(0.82)

0.332

)

-0.49

-0.05

91.0

(0.82)

0.101

2)

GAPDH HPRT1

0.144

3.19

3.77

0.05*

3.64

3.77

0.680

3.42

3.77

0.194

ACCEPTED MANUSCRIPT + TBP

(0.96

(1.22

(0.96

+ 18S

)

)

)

GAPDH

5.65

6.19

+ TBP

(0.99

+ 18S

(1.22

(0.62)

) 6.29

0.048

6.19

6.29

(0.97

(0.99

(0.48)

*

(0.97

(0.48)

)

)

)

HPRT1

3.45

4.02

+ TBP

(1.12

+

)

0.322

4.13

0.045

4.02

4.13

(0.89

(1.12

(0.62)

*

(0.89

(0.62)

)

)

)

All

3.81

4.43

possible

(1.30

referenc

)

3.81

4.44

0.045

(1.41

(1.30

(0.57)

)

)

)

NU

+ 18S 0.144

B2M +

9.98

10.5

0.049

9.98

18Sb

(0.92

2

*

(0.92

)

(1.42

(1.41

4.44

0.68

(0.57)

MA (0.72)

0.077

6.03

6.29

(0.94

(0.48)

0.101

) 3.81

4.13

0.011

(0.97

(0.62)

9

3.95

4.44

0.186

(1.41

(0.57)

)

)

10.5

10.62

2

(0.72)

0.804

10.2

10.62

0

(0.72)

(1.42

(1.34

)

)

0.406

A lower cycle threshold value (Crt) indicates higher gene expression; bB2M + 18S was

PT

a

10.62

ED

)

*

4.43

)

e genes

)

0.409

SC RI

3.45

GAPDH

(0.62)

)

5.65

0.136

(1.21

PT

0.152

(0.62)

used as an example of not suitable reference gene pair for gene expression

CE

normalization in meniscus samples. *p ≤ 0.05 by Mann-Whitney test. IQR: interquartile

AC

range; MI: isolated meniscal tear samples; ML: meniscus samples of patients with concomitant meniscal tear and anterior cruciate injury; MC: meniscus samples of controls.

ACCEPTED MANUSCRIPT Highlights 

Stable reference genes are necessary to a reliable expression data by quantitative



PT

PCR. We determined a set of reference genes for qPCR use in injured and non-injured



SC RI

menisci.

The stability of reference genes were assessed using NormFinder, geNorm, BestKeeper DataAssist and RefFinder software packages and comparative ΔCt.

MA

HPRT1+TBP+GAPDH is the most suitable combination for the analysis of

CE

PT

ED

injured and non-injured menisci.

AC



NU

method.