HomeLarge Type Edition
HOME ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 61:20-27 (2006)
© 2006 The Gerontological Society of America

Identification of Valid Housekeeping Genes and Antioxidant Enzyme Gene Expression Change in the Aging Rat Liver

Jie Chen1,2, David A. Rider2 and Runsheng Ruan1,2,

1 Department of Otolaryngology, National University Hospital, Singapore.
2 Cancer and Ageing Research, Institute of Bioengineering and Nanotechnology, Singapore.

Address correspondence to Runsheng Ruan, MD, Institute of Bioengineering and Nanotechnology, 31 Biopolis Way, The Nanos, #04-01, Singapore 138669. E-mail: rsruan{at}ibn.a-star.edu.sg


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 References
 
Valid housekeeping genes (HKG) are a prerequisite for accurate gene quantification. We performed real-time reverse transcription–polymerase chain reaction to investigate the gene expression of five commonly used HKGs (ß-actin, glyceraldehyde-3-phosphate dehydrogenase [GAPDH], ubiquitin C [UBC], hypoxanthine phosphoribosyl-transferase [HPRT], and cyclophilin A [CYPa]) and antioxidant enzymes in the liver of young and old male Fischer rats. A wide variation in HKG expression existed during the aging process, and HPRT was identified as the most stable HKG in rat liver aging. When Cu/Zn-superoxide dismutase gene expression was normalized to HPRT, there was no detectable difference between young and old rats; however, a significant difference was seen when it was normalized to UBC. The variation of UBC caused the misinterpretation of Cu/Zn-superoxide dismutase expression. Catalase expression was significantly decreased, whereas glutathione peroxidase expression was not altered with age. We demonstrated that HPRT was an appropriate HKG, validation of HKGs was vital for accurate quantification, and decreased catalase expression might be involved in the decline of antioxidant defenses during rat liver aging.

THE use of gene expression profiling techniques is becoming an integral part of research aiming to fully understand the complex mechanisms of aging. The development of microarray technology has provided researchers with a powerful tool to do this (1,2). However, this technology has also posed a number of problems in interpretation due to difficulties in validation of the data (3,4). In small-scale studies, the detection of RNA can be accomplished by common techniques including northern blot analysis, RNase protection assays, competitive reverse transcription–polymerase chain reaction (RT–PCR), and real-time RT–PCR. Among them, real-time RT–PCR has become increasingly popular compared to other methods because it has higher sensitivity, greater speed, and broader dynamic quantification range (5,6).

However, for all these methods, normalization is required to correct for variation in RNA integrity, RT efficiency, and initial sample amount among different samples. For normalization, housekeeping genes (HKGs) are widely used; the stability of the HKGs is a prerequisite for accurate normalization. Unfortunately, the expression of these assumed stable HKGs, such as ß-actin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), has been reported to vary greatly in different tissues and under different experimental conditions, making the selection of a "universal" HKG for all experiments problematic (7,8). Furthermore, poor selection of HKGs can invalidate the normalization process and lead to the generation of misleading information (9,10). For example, in a study of human clear cell renal cell carcinoma (CCRCC), when the target gene, tumor suppressor gene P53 (TP53), was normalized to the selected HKG, lamin B1 (LMNB1), a significant difference was detected between normal and CCRCC groups. In fact, it was not TP53, but LMNB1 gene expression that changed during that study (10). Thus, this has led to the proposal that appropriate HKGs should be validated in each specific experiment to provide reliable normalization of data (11–13).

Some software packages, such as GeNorm (11) and Normfinder (14), have made the process of HKG validation easier. In addition, the data from real-time RT–PCR can be effectively analyzed by using statistics, thus the objectivity of the result is improved. The GeNorm software is regarded as the authoritative method for analysis (15) and is consequently the most popular method used. It chooses appropriate HKGs by pair-wise comparison between one HKG and all other HKGs independent of the level of gene expression, and does not require a normal distribution of data. However, co-regulation of HKGs will influence the efficiency of this method because of the pair-wise comparison. Trying to predict the co-regulation of HKGs is difficult because in addition to their basic roles, some HKGs also have other diverse functions (16,17). The alternative software Normfinder ranks HKGs according to the least estimated intra- and intergroup variation, which is more effective to control the influence of co-regulation of HKGs. Thus, both methods were used in this study.

In aging research, real-time RT–PCR has been increasingly used to detect changes in gene expression (18–20). However, as yet there has not been a study to identify valid HKGs to use in aging research. In the present study, we used real-time RT–PCR to study the expression of five commonly used HKGs: ß-actin, GAPDH, ubiquitin C (UBC), hypoxanthine phosphoribosyl-transferase (HPRT), and cyclophilin A (CYPa) from the liver of young and old Fischer rats and chose a valid HKG using the HKG identification softwares GeNorm and Normfinder. Under certain experimental conditions and disease states, the expression of some HKGs can vary greatly, whereas other HKGs remain relatively stable (7,8). The five chosen HKGs were selected as they all have different physiological functions—cytoskeleton (ß-actin), carbohydrate metabolism (GAPDH), protein folding (CYPa), metabolic salvaging of nucleotides (HPRT), and protein degradation (UBC)—thus minimizing the risk that the aging process would affect all of the genes. Each of the genes studied here has been recommended as a suitable HKG in at least one biological condition (21–25). In addition, to illustrate the significance of choosing appropriate HKGs, normalization of Cu/Zn-superoxide dismutase (Cu/Zn-SOD) gene expression to the different HKGs was performed. Finally, it has been widely accepted that oxidative damage plays an important role in the aging process [recently reviewed in (26)], and age-related antioxidant enzyme activities have been intensively studied (27), but the effect of age on the gene expression of antioxidant enzymes has not been extensively studied and findings are contradictory (28,29). Therefore, we also investigated the gene expression levels of catalase and glutathione peroxidase (GPX), two other important antioxidant enzymes, in the aged rat liver by real-time RT–PCR.


   METHODS
Top
 Abstract
 Methods
Results
 Discussion
 References
 
Animals and Harvesting Liver Tissue
Young (8 months, n = 9) and old (26 months, n = 8) male specific pathogen-free Fischer 344 rats were obtained from the National Institute on Aging (Bethesda, MD), and the livers were removed after the animals were killed. Liver tissues were immersed in RNALater (Ambion, Austin, TX) overnight at 4°C before storage at –20°C until analysis. All animal procedures were approved by the Institutional Animal Care and Use Committee of the National University of Singapore.

RNA Isolation, DNase Treatment of RNA, and RT
Total RNA was isolated from liver tissue using the RNeasy Mini Kit (Qiagen, Hilden, Germany), according to the manufacturer's instructions. To avoid contamination with genomic DNA, total RNA was DNase-treated both on-column (RNase-free DNase set; Qiagen) and off-column (RNase free DNase I Kit; Invitrogen, Singapore) according to the manufacturer's instructions. Absence of genomic DNA was confirmed by conventional PCR using primers designed to recognize the first intron of ß-tubulin (forward primer 5'-acactttctgtccgctcacac-3' and reverse primer 5'-gtgcgcattacctcaacaaag-3'). The concentration of total RNA was quantified using a spectrophotometer at 260 nm, with the optical density (OD)260/OD280 ratio routinely between 2.00 and 2.12. RNA integrity was confirmed by formaldehyde gel electrophoresis (as described in the manual of Qiagen RNeasy Mini Kit). First-strand complementary DNA was synthesized using the SuperScript III First-Strand Synthesis System (Invitrogen), according to the manufacturer's instructions with Oligo(dT)20 and 600 ng of RNA. One sample without RT was included as the negative control.

Optimization of PCR and Real-Time PCR
Optimization was carried out using the QuantiTect SYBR Green PCR Kit (Qiagen), using the gradient procedure of a DNA Engine Opticon 2 (MJ Research, Waltham, MA). The specificity of PCR products was confirmed by both melting curve analysis and agarose gel electrophoresis.

Real-time PCR was performed under optimal conditions using the following PCR amplification mixture (20 µl total): 2x QuantiTect SYBR Green PCR master mix, 0.3 µM forward and reverse primers (Table 1), and 0.8 µl of 1:10 diluted complementary DNA. The cycling conditions were as follows: 15-minute initial activation step at 95°C; then 94°C for 15 seconds, annealing at optimal temperature for 30 seconds, and 72°C for 30 seconds (repeated for 40 cycles); finally the melting curve analysis was performed, and samples were then cooled to 10°C. Fluorescent data was acquired at high temperatures to avoid inference of nonspecific fluorescence signals (34). Each assay included a no-template control and an RT negative control. For each gene tested, all samples were detected in triplicate in the same plate, avoiding interplate variation. A relative standard curve was produced from a 3-fold dilution series across five data points to calculate amplification efficiencies and correlation coefficients (R2). Amplification efficiencies were calculated using the equation: E = 10[–1/slope] – 1. Data analysis was performed according to the manual of GeNorm 3.4 (http://medgen.ugent.be/%7Ejvdesomp/genorm/) (11) and Normfinder (http://www.mdl.dk/publications_normfinder.htm) (14), respectively.


View this table:
[in this window]
[in a new window]
  		Table 1. PCR Primers, PCR Efficiency, and Correlation Coefficient.

 
HKG Stability Analysis
GeNorm 3.4 and Normfinder were used to evaluate HKG stability. GeNorm ranked gene stability by an average expression stability value (M), which was the average pair-wise variation of an HKG compared with all other HKGs. Normfinder ranked gene stability by stability values which were derived from intra-group variation and intergroup variation. More stable gene expression was indicated by lower average expression stability values.

Normalization Factor Determination by GeNorm
The normalization factor (NF) was calculated from multiple HKGs and was thought to be a more accurate method of normalization. In our experiment, an NF from HPRT–GAPDH–ß-actin was determined by GeNorm. The theory of choosing and calculating NF by GeNorm was described in Vandesompele and colleagues (11). Briefly, from the stability analysis using GeNorm, it was found that the two most stable HKGs in our experiment were HPRT and GAPDH. NF2 was based on the expression levels of these two genes and was calculated using different combinations of genes and including the most stable remaining control gene (HPRT–GAPDH–ß-actin [NF3], HPRT–GAPDH–ß-actin–CYPa [NF4], and HPRT–GAPDH–ß-actin–CYPa–UBC [NF5]). For every series of NFn and NFn+1, pair-wise variation was calculated; for example, NF2:NF3, then NF3:NF4, and finally NF4:NF5. A high pair-wise variation value meant that the added gene had a significant effect on normalization and should be included in the NF. According to Vandesompele and colleagues (11), the ideal pair-wise variation value is less than 0.15. Thus, following the pair-wise variation analysis, it was decided that the NF would be derived from HPRT–GAPDH–ß-actin in our experiment.

Statistical Analysis
Mann–Whitney U nonparametric analysis was performed to detect significant differences among young and old groups. Values of p <.05 were considered significant differences.


   RESULTS
Top
 Abstract
 Methods
 Results
Discussion
 References
 
Real-Time RT-PCR Specificity, Efficiency, and Linearity
Following real-time RT-PCR, the resulting PCR products were analyzed in two ways to confirm the specificity of the reaction. The PCR products from each gene of interest were analyzed by agarose gel electrophoresis (data not shown) and melting curve analysis as shown in Figure 1, A (Cu/Zn-SOD) and B (HPRT). Only a single PCR product is evident via these two analysis methods, confirming the specificity of the PCR. A relative standard curve method was used to calculate amplification efficiencies and R2 values. All tested gene PCR efficiencies were above 96%, and correlation coefficients were more than 0.99 (Table 1). Thus, the data obtained from the real-time RT–PCR experiments were reliable and could be used for further analysis.


Figure 01
View larger version (19K):
[in this window]
[in a new window]
  		Figure 1. The real-time reverse transcription–polymerase chain reaction (RT–PCR) was specific. In real-time RT–PCR, a melting curve analysis was performed to demonstrate the specificity of the reactions. Plots show melting curves of Cu/Zn-superoxide dismutase (A) and hypoxanthine phosphoribosyl-transferase (B). Only a single peak is evident in the melting curve, demonstrating the high specificity of the reaction

 
The Stability Sequence of HKGs
To identify the HKG with the most stable expression in the rat liver aging process, the HKG selection software packages GeNorm and Normfinder were used. HPRT was identified as the most stable HKG by both analysis packages. GeNorm identified GAPDH and HPRT as the most stable HKGs (Figure 2), with stability values of 0.42 for both genes. When the analysis from Normfinder was also taken into account (Table 2), we were able to select HPRT over GAPDH as our HKG because the intra-group variation was very small for HPRT compared to GAPDH (0.010 compared to 0.064 for young rats and 0.018 compared to 0.089 for old rats), and the intergroup variations were comparable (0.040 for HPRT, 0.038 for GAPDH).


Figure 02
View larger version (8K):
[in this window]
[in a new window]
  		Figure 2. Hypoxanthine phosphoribosyl-transferase (HPRT) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were the most stable housekeeping genes (HKGs) by GeNorm. GeNorm software was used to identify the most stable HKG using a pair-wise variation analysis. Plot shows the average expression stability values for the HKGs, with a low value representing stable HKG expression. UBC = ubiquitin C; CYPa = cyclophilin A

 

View this table:
[in this window]
[in a new window]
  		Table 2. Expression Stability of Housekeeping Genes Evaluated by Normfinder.

 
Both programs indicated that a wide variation in HKG expression between young and old rats existed in the liver during the aging processes (Figure 2 and Table 2). After normalization to the confirmed HKG, HPRT, compared to young rats, messenger RNA (mRNA) expression of UBC was significantly decreased in old rats (p <.01), whereas mRNA expression of CYPa was significantly increased in old rats (p <.01) (Figure 3). There were no significant differences in GAPDH and ß-actin between the two groups, even though ß-actin mRNA in old rats increased by 29.6% compared to young rats (p =.07). These data have thus identified HPRT as the HKG of choice in rat liver aging studies.


Figure 03
View larger version (13K):
[in this window]
[in a new window]
  		Figure 3. Housekeeping gene (HKG) expression variation between young and old rats. To compare the HKG's expression variation between young and old rats, each HKG's expression was normalized to the confirmed HKG, hypoxanthine phosphoribosyl-transferase. Graph represents the average HKG expression in the liver of young (white bar, n = 9) and old (black bar, n = 8) rats ± standard error. *p <.01 as determined by the Mann–Whitney U test. mRNA = messenger RNA; UBC = ubiquitin C; CYPa = cyclophilin A

 
To clearly explain the HKG expression variability, we analyzed the relative expression ratios of the most or least stable HKG compared to other HKGs, according to the method used by Biederman and colleagues (35). As shown in Table 3, the relative expression ratio of two robust HKGs, HPRT and GAPDH, showed only a 0.28% difference, whereas the relative expression of HPRT and UBC demonstrated a 58.61% difference. This underlies the extent of the variation in the expression of the various HKGs.


View this table:
[in this window]
[in a new window]
  		Table 3. Relative Expression Ratios for the Best and Worst Housekeeping Gene (HKG), as Compared to the Other HKG Tested*.

 
Interpretation of Cu/Zn-SOD and Catalase Gene Expression Normalization by Different HKGs
To demonstrate the importance of selecting the correct HKG when using gene expression analysis techniques, we used different HKGs to compare the gene expression of Cu/Zn-SOD in liver tissue from young and old rats. As shown in Figure 4, when expression levels were normalized using UBC, an invalidated HKG, Cu/Zn-SOD expression had a significant difference in expression levels in liver tissue between young and old rats. On the contrary, after normalization to HPRT, the validated HKG, there was no statistically significant difference in Cu/Zn-SOD expression. An alternative method to quantify gene expression used the NF, which is based on the expression level of multiple HKGs. The GeNorm analysis software was used to calculate the NF in our samples, and then an NF value derived from HPRT–GAPDH–ß-actin expression was selected. When Cu/Zn-SOD expression was normalized using the NF, no statistically significant difference was seen (Figure 4). In addition, there was a statistically significant difference in catalase gene expression when it was normalized to HPRT (p <.001) and UBC (p <.05), yet it should be noted that the level of catalase mRNA decreased by only 31.20% when normalized using UBC (this is about half the value when normalized using HPRT [57.73%]). These data demonstrated that selection of a validated HKG was essential for the correct interpretation of gene expression analysis data in aging research.


Figure 04
View larger version (10K):
[in this window]
[in a new window]
  		Figure 4. Influence of different housekeeping genes on interpretation of Cu/Zn-superoxide dismutase (Cu/Zn-SOD) messenger RNA (mRNA) expression in the rat liver aging process. The level of Cu/Zn-SOD mRNA was normalized to ubiquitin C (UBC) (first two bars), hypoxanthine phosphoribosyl-transferase (HPRT) (middle two bars), and normalization factor (NF) (last two bars), respectively. Data were expressed as mean ± standard error. Young rats (white bar, n = 9) and old rats (black bar, n = 8). *p <.05 as determined by the Mann–Whitney U test

 
Antioxidant Enzyme Gene Expression Changes in the Aged Rat Liver
To further analyze antioxidant enzyme expression in the aged rat liver, we measured catalase and GPX expression. As shown in Figure 5, mRNA expression of catalase decreased significantly in liver tissue between young and old rats, with expression levels in the old rats only 42.27% of that seen in the young rats. Similar to the results for Cu/Zn-SOD (Figure 4), there was no significant difference in GPX expression; however, there did appear to be a trend for increased expression of GPX in the elderly rats, although it did not reach significance (p =.06).


Figure 05
View larger version (8K):
[in this window]
[in a new window]
  		Figure 5. Antioxidant enzyme expression variation between young and old rats. To compare antioxidant enzyme expression variation between young and old rats, each antioxidant enzyme expression was normalized to the confirmed housekeeping gene, hypoxanthine phosphoribosyl-transferase. Graph represents the antioxidant enzyme gene expression in the liver of young (white bar, n = 9) and old (black bar, n = 8) rats ± standard error. *p <.001 as determined by the Mann–Whitney U test. mRNA = messenger RNA; GPX = glutathione peroxidase

 

   DISCUSSION
Top
 Abstract
 Methods
 Results
 Discussion
References
 
We have identified HPRT as the most stable HKG to use for studying rat liver aging using the software GeNorm and Normfinder. Analysis using GeNorm identified HPRT and GAPDH as the two most stable HKGs (Figure 2), but could not differentiate further between them because the use of gene ratios was needed for gene stability measurements. When using Normfinder, HPRT had more stable intra-group variation and had similar inter-group variation when compared to GAPDH (Table 2). Thus, HPRT was chosen as the most stable HKG. Recently, HPRT had also been recommended as the best HKG to use in human cancer research (36). Previously GeNorm has been used to select HKGs for use in the study of human liver diseases (25), leukocytes and bone marrow (11), with UBC identified as the most stable HKG. In contrast to those studies, UBC was ranked as the most unstable HKG by both methods in this present study of rat liver aging (see Figure 2 and Table 2). These results demonstrate that the appropriate HKG to use should be validated according to the specific tissue and specific experimental conditions, as has been previously suggested (11,13,37).

Although aging is a physiological process, we found that there was a wide variation in HKG expression during rat liver aging. Previous studies have identified obvious HKG variations which have been attributed to pathological changes (12,13,36) or cell differentiation (37,38). Two commonly used HKGs, UBC and CYPa, had significantly different expression levels between young and old rats (Figure 3). In addition, ß-actin mRNA level in the liver of old rats increased by 29.6%, although this finding did not reach statistical significance. Similarly, Moshier and colleagues (39) found that ß-actin expression was also different in rat gastric mucosa, where there was a decrease in expression by 37% in old rats (24 months) compared to young rats (6 months) using northern blot analysis. The difference in ß-actin expression between our study and the study by Moshier and colleagues may be attributed to tissue-specific HKG variation, as has been discussed above. However, the validation of the HKG could not be further improved at that time because of the lack of a powerful statistical method and the decreased sensitivity of northern blot analysis compared with real-time RT–PCR. In other words, HKG expression variability could be revealed more clearly by analyzing the relative expression ratios of two HKGs. For two ideal HKGs, the relative expression ratios between young and old rats should be similar. In our study, the relative expression ratio of HPRT and GAPDH showed a 0.28% difference, whereas the relative expression ratio of HPRT and UBC showed a 58.61% difference. This variation means that if gene expression was normalized to UBC, the target gene expression would be overestimated by 58.61%.

In addition, HKG variation could cause confusing, even misleading interpretation of gene expression data. Normalization to the confirmed HKG, HPRT, did not identify any significant difference in Cu/Zn-SOD expression during rat liver aging processes, whereas normalization to the invalidated HKG, UBC, demonstrated significant differences in the same experiment (Figure 4). This finding was attributed to the variation in expression of the HKG, and thus in this experiment it was UBC rather than Cu/Zn-SOD that was fluctuating in the aging process. Similar cases have been reported in human asthma (9) and renal carcinoma (10). Even though normalization to different HKGs did not influence the findings for catalase expression, the quantification of catalase expression change was distorted. When normalized to HPRT, catalase gene expression in the aged rats decreased by 57.73% compared to 31.20% when normalized to the invalidated UBC gene. From current data, most of the age-related gene expression changes occurred in the 30% to 3-fold range (40,41). Therefore, validation of HKGs in aging research is absolutely vital for accurate gene expression quantification.

The use of more than one HKG for normalization has been proposed because of the obvious HKG expression variations in some experiments (11). In those cases, the NF calculated from several HKGs may bring more accurate normalization. In the present study, an NF calculated from HPRT–GAPDH–ß-actin was recommended for gene expression analysis in rat liver aging processes by GeNorm software. However, the feasibility of using NF has been argued (13,36), especially in studies where there is limited RNA. It is interesting that normalization of Cu/Zn-SOD expression to the NF or to HPRT alone produced similar results in this study. Therefore, HPRT was a suitable HKG to use for the accurate normalization of gene expression in rat liver aging research.

In previous studies using northern blot analysis, an age-dependent decline in expression of antioxidant enzymes was found, and the extent of these enzymes' expressional change was less than 3-fold (28,42). The limited sensitivity of northern blot analysis might not be suitable for identifying so small an expression change, thus those results need to be further investigated. However, real-time RT–PCR is believed to determine accurately changes as small as 2-fold using statistical methods (43). In addition, it was reported that more than 90% of age-related gene expression changes were less than 3-fold in magnitude (40,41). Therefore, the use of real-time RT–PCR with valid HKGs will be a reasonable and practical tool for detecting age-related gene expression alterations.

In our study, the expression of catalase was significantly decreased, whereas Cu/Zn-SOD and GPX expression did not show a significant change between young and old male Fischer 344 rats. Studies using northern blot analysis showed decreasing expression of Cu/Zn-SOD and catalase and no change in GPX expression with age in male Fischer 344 rats (28,42), whereas a study using common RT–PCR showed no change in Cu/Zn-SOD with age in male Fischer 344 rats (29). Thus, the investigations of age-related catalase and GPX gene expression were relatively consistent, and the discrepancy in Cu/Zn-SOD gene expression might be partly from different experimental methodologies. It has been reported that changes in antioxidant enzyme activities during aging may be dependent on rat strain (44,45); yet age-related antioxidan enzyme gene expression changes presented here from male Fischer 344 rats were similar to those reported from male Wistar rats (46), whereas a study using northern blot analysis in Wistar rats showed increasing expression of Cu/Zn-SOD, catalase, and GPX expression from 6 to 30 months of age (47). Real-time RT–PCR was also used in that study but without the use of a validated HKG, the similarity in antioxidant enzyme gene expression in the two rat strains would still need further validation. In addition, because Cu/Zn-SOD, catalase, and GPX are major antioxidant enzymes in eukaryotes and the decrease of antioxidant defense had been widely demonstrated during rat liver aging (27,28,48), the decrease in catalase gene expression might be one reason of the diminution of the intricate antioxidant defense system. GPX mRNA expression seemed to increase, even though it did not reach significance (p =.06). The trend for increased expression of GPX may compensate for the decreased expression of catalase because catalase and GPX cooperate to remove hydrogen peroxide. This inference needs to be further investigated. Our data showed that catalase mRNA expression decreased during the rat liver aging process, which might be involved in the decline of the intricate antioxidant defense system.

Obtaining reliable RT–PCR data is a prerequisite for selecting appropriate HKGs to avoid nonspecific binding of SYBR Green I dye (49). A number of measures were taken to achieve the desired specificity of PCR product in this study. First, genomic DNA contamination was sufficiently controlled by DNase treatment both on-column and off-column, because traces of genomic DNA will obviously prevent accurate quantification, especially when studying genes with unknown intron/exon structure, or when pseudogenes exist. Peters and colleagues (50) exhaustively discussed the necessity of combining on-column and off-column DNase treatment in real-time RT–PCR and its impact on efficiency. Second, primers that flanked introns were used to ascertain the absence of genomic DNA because genomic DNA would have the intron sequence amplified. Third, the effective control of genomic DNA was confirmed by using primers designed to recognize the first intron of ß-tubulin in a conventional PCR before RT. Finally, the specificity of the PCR products was verified by both melting curve analysis and agarose gel electrophoresis. In addition, during the real-time RT–PCR, fluorescence data were acquired at a high temperature, avoiding the inference of nonspecific fluorescence signals (30). During data analysis, PCR efficiency-corrected quantification was performed because small PCR efficiency differences between target genes and the HKG will have an effect on the calculation of gene expression. For example, a 0.03% difference of efficiency will lead to an error in the gene expression ratio of 46% when Etarget < EHKG and 209% when Etarget EHKG after 25 PCR cycles (51).

Summary
This study demonstrated that there was a wide variation in HKG expression during aging. We have identified HPRT as an appropriate HKG to use for accurate normalization during gene expression analysis of the effects of aging on the rat liver. To our knowledge, this is the first report discussing validation of HKGs in aging research where choosing appropriate HKGs is vital for accurate gene quantification and analysis. Further investigation showed that catalase expression was significantly decreased, whereas Cu/Zn-SOD and GPX expression were not significantly altered with age; this result indicates that decreased catalase gene expression might be involved in the decline of the antioxidant defense system in the rat liver aging process.


   Acknowledgments
Top
 Abstract
 Methods
 Results
 Discussion
 References
 
This work was financially supported by the Institute of Bioengineering and Nanotechnology, the Agency for Science, Technology and Research (A* STAR), Singapore and the Biomedical Research Council (Project No. 01/1/21/19/172), Singapore.


   Footnotes
Top
 Abstract
 Methods
 Results
 Discussion
 References
 
Decision Editor: James R. Smith, PhD

Received May 9, 2005

Accepted July 29, 2005


   References
Top
 Abstract
 Methods
 Results
 Discussion
 References
 

  1. Iyer VR, Eisen MB, Ross DT, et al. The transcriptional program in the response of human fibroblasts to serum. Science. 1999;283:83-87.[Abstract/Free Full Text]
  2. Brown PO, Botstein DPO. Exploring the new world of the genome with DNA microarrays. Nat Genet. 1999;21:33-37.[Medline]
  3. Han E, Hilsenbeck SG. Array-based gene expression profiling to study aging. Mech Ageing Dev. 2001;122:999-1018.[Medline]
  4. Tan PK, Downey TJ, Spitznagel EL, Jr, et al. Evaluation of gene expression measurements from commercial microarray platforms. Nucleic Acids Res. 2003;31:5676-5684.[Abstract/Free Full Text]
  5. Ginzinger DG. Gene quantification using real-time quantitative PCR: an emerging technology hits the mainstream. Exp Hematol. 2002;30:503-512.[Medline]
  6. Bustin SA. Quantification of mRNA using real-time reverse transcription PCR (RT-PCR): trends and problems. J Mol Endocrinol. 2002;29:23-39.[Abstract]
  7. Thellin O, Zorzi W, Lakaye B, et al. Housekeeping genes as internal standards: use and limits. J Biotechnol. 1999;75:291-295.[Medline]
  8. Warrington JA, Nair A, Mahadevappa M, Tsyganskaya M. Comparison of human adult and fetal expression and identification of 535 housekeeping/maintenance genes. Physiol Genomics. 2000;2:143-147.[Abstract/Free Full Text]
  9. Glare EM, Divjak M, Bailey MJ, Walters EH. Beta-actin and GAPDH housekeeping gene expression in asthmatic airways is variable and not suitable for normalising mRNA levels. Thorax. 2002;57:765-770.[Abstract/Free Full Text]
  10. Haller F, Kulle B, Schwager S, et al. Equivalence test in quantitative reverse transcription polymerase chain reaction: confirmation of reference genes suitable for normalization. Anal Biochem. 2004;335:1-9.[Medline]
  11. Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol [serial online]. 2002; 3. Available at: http://genomebiology.com/2002/3/7/RESEARCH/0034. Accessed June 18, 2002.
  12. Aerts JL, Gonzales MI, Topalian SL. Selection of appropriate control genes to assess expression of tumor antigens using real-time RT-PCR. Biotechniques. 2004;36:84-91.[Medline]
  13. Dheda K, Huggett JF, Bustin SA, Johnson MA, Rook G, Zumla A. Validation of housekeeping genes for normalizing RNA expression in real-time PCR. Biotechniques. 2004;37:112-119.[Medline]
  14. Andersen CL, Jensen JL, Orntoft TF. 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. 2004;64:5245-5250.[Abstract/Free Full Text]
  15. Bustin SA, Nolan T. Pitfalls of quantitative real-time reverse-transcription polymerase chain reaction. J Biomol Tech. 2004;15:155-166.[Medline]
  16. Singh R, Green MR. Sequence-specific binding of transfer RNA by glyceraldehyde-3-phosphate dehydrogenase. Science. 1993;259:365-368.[Abstract/Free Full Text]
  17. Ishitani R, Sunaga K, Hirano A, Saunders P, Katsube N, Chuang DM. Evidence that glyceraldehyde-3-phosphate dehydrogenase is involved in age-induced apoptosis in mature cerebellar neurons in culture. J Neurochem. 1996;66:928-935.[Medline]
  18. Mohamed SA, Wesch D, Blumenthal A, et al. Detection of the 4977 bp deletion of mitochondrial DNA in different human blood cells. Exp Gerontol. 2004;39:181-188.[Medline]
  19. Masternak MM, Al-Regaiey K, Bonkowski MS, et al. Divergent effects of caloric restriction on gene expression in normal and long-lived mice. J Gerontol A Biol Sci Med Sci. 2004;59:784-788.
  20. Salles N, Szanto I, Herrmann F, et al. Expression of mRNA for ROS-generating NADPH oxidases in the aging stomach. Exp Gerontol. 2005;40:353-357.[Medline]
  21. Choi JK, Holtzer S, Chacko SA, Lin ZX, Hoffman RK, Holtzer H. Phorbol esters selectively and reversibly inhibit a subset of myofibrillar genes responsible for the ongoing differentiation program of chick skeletal myotubes. Mol Cell Biol. 1991;11:4473-4482.[Abstract/Free Full Text]
  22. Wall SJ, Edwards DR. Quantitative reverse transcription-polymerase chain reaction (RT-PCR): a comparison of primer-dropping, competitive, and real-time RT-PCRs. Anal Biochem. 2002;300:269-273.[Medline]
  23. Peinnequin A, Mouret C, Birot O, et al. Rat pro-inflammatory cytokine and cytokine related mRNA quantification by real-time polymerase chain reaction using SYBR green. BMC Immunol [serial online]. 2004; 5. Available at: http://www.biomedcentral.com/1471-2172/5/3. Accessed February 5, 2004.
  24. Foss DL, Baarsch MJ, Murtaugh MP. Regulation of hypoxanthine phosphoribosyltransferase, glyceraldehyde-3-phosphate dehydrogenase and beta-actin mRNA expression in porcine immune cells and tissues. Anim Biotechnol. 1998;9:67-78.[Medline]
  25. Kim S, Kim T. Selection of optimal internal controls for gene expression profiling of liver disease. Biotechniques. 2003;35:456-460.[Medline]
  26. Lombard DB, Chua KF, Mostoslavsky R, Franco S, Gostissa M, Alt FW. DNA repair, genome stability, and aging. Cell. 2005;120:497-512.[Medline]
  27. Grune T, Davies KJ. Oxidative processes in aging. In: Masoro EJ, Austad SN, eds. Handbook of the Biology of Aging. 5th ed. San Diego, CA: Academic Press; 2001:25–58.
  28. Rao G, Xia E, Richardson A. Effect of age on the expression of antioxidant enzymes in male Fisher F344 rats. Mech Ageing Dev. 1990;53:49-60.[Medline]
  29. Thomas RP, Guigneaux M, Wood T, Evers BM. Age-associated changes in gene expression patterns in the liver. J Gastrointest Surg. 2002;6:445-453.[Medline]
  30. Zhang AS, Xiong S, Tsukamoto H, Enns CA. Localization of iron metabolism-related mRNAs in rat liver indicate that HFE is expressed predominantly in hepatocytes. Blood. 2004;103:1509-1514.[Abstract/Free Full Text]
  31. Depreter M, Vandesompele J, Espeel M, Speleman F, Roels F. Modulation of the peroxisomal gene expression pattern by dehydroepiandrosterone and vitamin D: therapeutic implications. J Endocrinol. 2002;175:779-792.[Abstract]
  32. Noda M, Kariura Y, Amano T, et al. Expression and function of bradykinin receptors in microglia. Life Sci. 2003;72:1573-1581.[Medline]
  33. Song G, Cechvala C, Resnick DK, Dempsey RJ, Rao VL. GeneChip analysis after acute spinal cord injury in rat. J Neurochem. 2001;79:804-815.[Medline]
  34. Ball TB, Plummer FA, HayGlass KT. Improved mRNA quantitation in LightCycler RT-PCR. Int Arch Allergy Immunol. 2003;130:82-86.[Medline]
  35. Biederman J, Yee J, Cortes P. Validation of internal control genes for gene expression analysis in diabetic glomerulosclerosis. Kidney Int. 2004;66:2308-2314.[Medline]
  36. de Kok JB, Roelofs RW, Giesendorf BA, et al. Normalization of gene expression measurements in tumor tissues: comparison of 13 endogenous control genes. Lab Invest. 2005;85:154-159.[Medline]
  37. Hamalainen HK, Tubman JC, Vikman S, et al. Identification and validation of endogenous reference genes for expression profiling of T helper cell differentiation by quantitative real-time RT-PCR. Anal Biochem. 2001;299:63-70.[Medline]
  38. Bas A, Forsberg G, Hammarstrom S, Hammarstrom ML. Utility of the housekeeping genes 18S rRNA, beta-actin and glyceraldehyde-3-phosphate-dehydrogenase for normalization in real-time quantitative reverse transcriptase-polymerase chain reaction analysis of gene expression in human T lymphocytes. Scand J Immunol. 2004;59:566-573.[Medline]
  39. Moshier JA, Cornell T, Majumdar AP. Expression of protease genes in the gastric mucosa during aging. Exp Gerontol. 1993;28:249-258.[Medline]
  40. Pahlavani MA, Haley-Zitlin V, Richardson A. Influence of dietary restriction on gene expression: changes in the transcription of specific genes. In Yu BP, ed. Modulation of the Aging Process by Dietary Restriction. Boca Raton, FL: CRC Press; 1994:143–156.
  41. Van Remmen H, Ward W, Sabia RV, Richardson A. Effect of age on gene expression and protein degradation. In: Masoro EJ, ed. Handbook of Physiology, Volume on Aging. New York: Oxford University Press; 1995:171–234.
  42. Rao G, Xia E, Nadakavukaren MJ, Richardson A. Effect of dietary restriction on the age-dependent changes in the expression of antioxidant enzymes in rat liver. J Nutr. 1990;120:602-609.
  43. Walker NJ. A technique whose time has come. Science. 2002;296:557-558.[Abstract/Free Full Text]
  44. Rikans LE, Moore DR, Snowden CD. Sex-dependent differences in the effects of aging on antioxidant defense mechanisms of rat liver. Biochim Biophys Acta. 1991;1074:195-200.[Medline]
  45. Jang I, Chae K, Cho J. Effects of age and strain on small intestinal and hepatic antioxidant defense enzymes in Wistar and Fisher 344 rats. Mech Ageing Dev. 2001;122:561-570.[Medline]
  46. Martin R, Fitzl G, Mozet C, Martin H, Welt K, Wieland E. Effect of age and hypoxia/reoxygenation on mRNA expression of antioxidative enzymes in rat liver and kidneys. Exp Gerontol. 2002;37:1481-1487.[Medline]
  47. Sanz N, Diez-Fernandez C, Alvarez A, Cascales M. Age-dependent modifications in rat hepatocyte antioxidant defense systems. J Hepatol. 1997;27:525-534.[Medline]
  48. Tian L, Cai Q, Wei H. Alterations of antioxidant enzymes and oxidative damage to macromolecules in different organs of rats during aging. Free Radic Biol Med. 1998;24:1477-1484.[Medline]
  49. Wittwer CT, Herrmann MG, Moss AA, Rasmussen RP. Continuous fluorescence monitoring of rapid cycle DNA amplification. Biotechniques. 1997;22:130-138.[Medline]
  50. Peters IR, Helps CR, Hall EJ, Day MJ. Real-time RT-PCR: considerations for efficient and sensitive assay design. J Immunol Methods. 2004;286:203-217.[Medline]
  51. Pfaffl MW, Horgan GW, Dempfle L. Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res [serial online]. 2002; 30. Available at: http://nar.oxfordjournals.org.libproxy1.nus.edu.sg/cgi/content/full/30/9/e36. Acessed March 4, 2002.




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
PubMed
Right arrow PubMed Citation

HOME ARCHIVE SEARCH TABLE OF CONTENTS