**4. Gene expression profiles change related to aging**

It is particularly important to identify gene expression characteristics and variation of heterogeneous population of cells with age in whole blood.

Fractions of lymphocytes, monocytes, neutrophils, eosinophils, and basophils in white blood cells showed insignificant differences with age as a result of ANOVA analysis. This study attempted to identify characteristics of age-related gene expression by taking into account of change in the number of expressed genes by age and similarities of gene expression intensity between individuals.

## **4.1. Characteristics of study subjects**

Five males and five females of 12 week old Clawn miniature pigs were housed individually in cages of 1.5 m2 at the SPF facility of the breeder (Japan Farm Co., Ltd, Kagoshima, Japan) for 18 weeks. Mean body weights of males and females at the beginning of the experiment were 7.0 kg and 6.9 kg respectively. During this period, all animals were fed with 450g/day standard dry feed (Kodakara73, Marubeni Nisshin Feed Co., Ltd., Tokyo Japan) with free access to water. Fetuses were taken out from their mothers on days 77 to the 84 days of the pregnancy by a Caesarean section. The unborn baby's sex was determined based on the shape of the vulva.


Values are mean±SD. †*P* values were calculated using one-way factorial ANOVA.

**Table 1.** Subject body weight results doi:10.1371/journal.pone.0019761.t001

102 Blood Cell – An Overview of Studies in Hematology

**3. The advantage of using miniature pigs** 

the SGSC, its research environment has been enhanced [11].

heterogeneous population of cells with age in whole blood.

intensity between individuals.

**4.1. Characteristics of study subjects** 

**4. Gene expression profiles change related to aging** 

are essential issues.

tube such as PAXgene. This avoids the cell separation process after sampling and minimizes the possibility of RNA denaturation. Usually, peripheral blood mononuclear cells (PBMCs) separation employs the difference of specific gravity between other blood components, which should be followed immediately after the blood sampling. Such manipulation requires a skilled operator to reduce the influence of separation procedures on gene expression. Second, the whole blood is a heterogeneous population of lymphocytes (monocytes, T-cells, and B-cells), granulocytes (neutrophils, eosinophils, and basophils), and platelets. One can expect that representative subpopulations in white blood cells may vary depending on the health condition of an individual. When a great alteration occurs in some subpopulations, the whole blood may also depart from the normal state of its age, because whole blood is a heterogeneous mixture of such subpopulations. Therefore, identification of gene expression characteristics and age-related variation in subpopulations in whole blood

Pigs are a useful model animals of humans because they have similar anatomy and digestive physiology to human [5-6]. In particular, miniature pigs are easier to breed and handle than other nonprimates, making them an optimal species for preclinical test [7]. Moreover, blood samples can be taken repeatedly and human medical devices such as endoscopes and MRI and CT scanners are also applicable. These advantages increasingly allow miniature pigs for laboratory animals, with recent progress in upgraded supply systems. In spite of some large-scale microarray studies on pigs, only a limited amount of fundamental data is available for pigs compared to other laboratory species [8-9]. In September 2003, the Swine Genome Sequencing Consortium (SGSC) was formed by industry, government, and academia, to promote pig genome sequencing under international coordination [10]. In November 2009, since the announcement of completed swine genome map by members of

It is particularly important to identify gene expression characteristics and variation of

Fractions of lymphocytes, monocytes, neutrophils, eosinophils, and basophils in white blood cells showed insignificant differences with age as a result of ANOVA analysis. This study attempted to identify characteristics of age-related gene expression by taking into account of change in the number of expressed genes by age and similarities of gene expression

Five males and five females of 12 week old Clawn miniature pigs were housed individually in cages of 1.5 m2 at the SPF facility of the breeder (Japan Farm Co., Ltd, Kagoshima, Japan) All blood samples were collected from the superior vena cava at 12, 16, 20, 24, and 30 weeks of age. Blood (EDTA), plasma (EDTA) and serum samples for hematology and biochemical tests were collected 24 hours after fasting. Hematology and biochemical tests were conducted by Clinical Pathology Laboratory, Inc. (http://www.patho.co.jp/index.html) (Kagoshima, Japan) using standard clinical methods.

Body weight change and hematological variation during breeding period are shown Table 1 and Table 2, respectively. One-way ANOVA analysis for age-related variations in red blood cell count (RBC), hemoglobin concentration (HGB), and hematocrit value (HCT) showed significant differences for both males and females. However, the mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC) remained unchanged. Differences in platelet count (PLT) and fibrinogen level (Fbg) were significant only for females. Any significant differences were not observed for both males and females for Prothrombin time (PT), activated partial thromboplastin time (ATPP), and the white blood cell count (WBC). Similarly to humans, the ratio of lymphocytes to white blood cells increased with maturation from 16 to 30 weeks of age. However, its difference was statistically insignificant according to ANOVA analysis. From 12 to 30 weeks of age, the ratios of granulocytes (neutrophils, eosinophils, and basophils), lymphocytes, and monocytes to white blood cells were unchanged, and differences were also insignificant.

## **4.2. Microarray gene expression profiles - Number of expressed genes**

To characterize the age-related gene expression in whole blood from miniature pigs, RNA analysis was conducted on bloods sampled from fetal stage, 12, 20, and 30 weeks subjects. Each RNA sample was analyzed by an Agilent #G2519F#20109 Porcine Gene Expression Microarray (44K) consisting of 43603 oligonucleotide probes.

The change in the number of expressed genes to identify age-related characteristics was examined. Microarray gene expressions were divided into two groups; "absent" and

j gy

"present", using flag indicators given by the scanner. Background level was determined from spot intensities outside the gene probing area. "Absent" was assigned to the spots whose intensities were less than the background level, while the rests were marked as "present." Then each gene was judged as either "expressed" or "unexpressed" based on the number of "present" events. We defined a certain gene as "expressed" when "present" exceeds 75% out of replicated events. A threshold of 75% was chosen by considering experimental deviation.

Whole Blood RNA Analysis, Aging and Disease 105

The number of expressed genes was less in fetal stage and infancy period but increased with age, reaching a steady state of gene expression after 20 weeks of age (Figure 1). Expressed genes for male and female were analyzed by one-way factorial ANOVA. Then Tukey-Kramer's method was applied only to significant groups. Differences between age groups (fetal stage, 12, 20, and 30 weeks of age) were significant for male, female, and mixed subjects of male and female. A Tukey-Kramer's multiple comparisons test revealed that differences between fetal stage and other age groups were statistically significant (p<0.001) for both male and female. Also, differences were significant (P<0.05) between 12 and 30

**Figure 1.** Number of genes expressed in whole blood of miniature pigs at different ages. In the graph,

**Age (weeks)**

12 20 30

Male Female

Variations in correlation coefficients among individuals of the same age and different age groups were evaluated. Pearson correlation coefficient was used for correlation analysis. Correlation coefficients for a total of 31 microarrays were obtained in normalized signals log-scale after excluding "absent" spots. A color-coded pairwise correlation matrix is shown

The average correlation coefficient within the same age group is shown in Figure 3. Variations in gene expression were greater for younger subjects, but it diminished with age while generating resembling expression patterns. Correlation coefficient within 30 weeks age group was slightly smaller than that within 20 weeks age group. However, this difference is smaller than other distant age groups. Significant differences were observed between any age groups according to an ANOVA analysis using Fisher's Z-transform. The average correlation coefficient between different age groups is shown in Figure 4. Significant

**4.3. Microarray gene expression profiles – Correlation of gene expression** 

in Figure 2. The color scale at the bottom indicates correlation strength.

represents male and represents female. Values are means±SD.

Fetal stage

0

5000

10000

**Number of expressed genes**

15000

20000

25000

doi:10.1371/journal.pone.0019761.g001

weeks females.


Biochemical variables for miniature pigs during the experiment are shown. Values are mean±SD. RBC, red blood cell count; HGB, hemoglobin concentration; HCT, hematocrit value; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin ; MCHC, mean corpuscular hemoglobin concentration; PLT, blood platelet count; PT, prothrombin time; ATPP, activated partial thromboplastin time; Fbg, fibrinogen level; WBC, white blood cell count; and NS: not significant. †*P* values were calculated using one-way factorial ANOVA.

**Table 2.** Subject hematology results doi:10.1371/journal.pone.0019761.t002 The number of expressed genes was less in fetal stage and infancy period but increased with age, reaching a steady state of gene expression after 20 weeks of age (Figure 1). Expressed genes for male and female were analyzed by one-way factorial ANOVA. Then Tukey-Kramer's method was applied only to significant groups. Differences between age groups (fetal stage, 12, 20, and 30 weeks of age) were significant for male, female, and mixed subjects of male and female. A Tukey-Kramer's multiple comparisons test revealed that differences between fetal stage and other age groups were statistically significant (p<0.001) for both male and female. Also, differences were significant (P<0.05) between 12 and 30 weeks females.

104 Blood Cell – An Overview of Studies in Hematology

j gy

experimental deviation.

**Hematological** 

**RBC, 104**

**PLT, 10<sup>4</sup>**

**WBC, 102**

**Table 2.** Subject hematology results doi:10.1371/journal.pone.0019761.t002

"present", using flag indicators given by the scanner. Background level was determined from spot intensities outside the gene probing area. "Absent" was assigned to the spots whose intensities were less than the background level, while the rests were marked as "present." Then each gene was judged as either "expressed" or "unexpressed" based on the number of "present" events. We defined a certain gene as "expressed" when "present" exceeds 75% out of replicated events. A threshold of 75% was chosen by considering

**analysis Sex n 12 weeks 16 weeks 20 weeks 24 weeks 30 weeks** *<sup>P</sup>* †

**HGB, g/dL** Male 5 14.9 ± 1.6 16.4 ± 1.2 17.3 ± 0.6 18.3 ± 0.4 17.7 ± 0.3 < .001

**HCT, %** Male 5 50.9 ± 5.1 53.6 ± 2.7 54.7 ± 2.1 58.4 ± 2.8 55.3 ± 1.2 < .05

**MCV, fL** Male 5 65.8 ± 1.0 66.3 ± 2.5 67.3 ± 2.9 65.1 ± 1.4 65.8 ± 2.2 NS **MCH, Pg** Male 5 19.8 ± 0.5 20.0 ± 1.1 20.1 ± 0.9 20.5 ± 0.6 20.6 ± 0.8 NS **CHC, %** Male 5 30.1 ± 0.4 30.2 ± 1.0 29.9 ± 0.9 31.5 ± 0.9 31.2 ± 0.8 NS

**PT, sec** Male 5 13.8 ± 3.2 15.5 ± 0.3 16.5 ± 0.9 15.9 ± 0.7 16.1 ± 0.6 NS

Female 5 < 20 < 20 < 20 < 20 < 20 **Fbg, mg/dl** Male 5 171.3 ±36.9 185.8 ± 93.8 169.4 ± 39.4 158.6 ± 9.0 147.8 ± 34.2 NS

**/µL** Male 5 62.0 ± 18.7 86.6 ±12.7 78.8 ± 24.7 79.6± 24.0 71.8 ± 13.2 NS Female 5 66.0 ± 23.4 74.0 ± 13.7 78.0 ± 18.7 72.4 ± 10.4 61.8 ± 11.3 NS

**Lymphocyte, %** Male 5 34.8 ± 12.1 45.2 ± 7.4 44.6 ± 9.3 36.8 ± 6.9 33.6 ± 7.6 NS **Neutrophil, %** Male 5 55.0 ± 10.9 43.1 ± 10.3 44.8 ± 7.4 52.2 ± 7.0 56.2 ± 9.2 NS **Eosinophil, %** Male 5 3.8 ± 2.2 3.1 ± 1.4 3.0 ± 1.9 5.0 ± 2.7 4.6 ± 1.7 NS **Basophil, %** Male 5 0.3 ± 0.5 0.3 ± 0.4 0.2 ± 0.4 0.0 ± 0.0 0.2 ± 0.4 NS **Monocyte, %** Male 5 6.3 ± 1.0 8.0 ± 3.2 7.4 ± 1.5 6.0 ± 2.1 5.4 ± 1.3 NS Biochemical variables for miniature pigs during the experiment are shown. Values are mean±SD. RBC, red blood cell count; HGB, hemoglobin concentration; HCT, hematocrit value; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin ; MCHC, mean corpuscular hemoglobin concentration; PLT, blood platelet count; PT, prothrombin time; ATPP, activated partial thromboplastin time; Fbg, fibrinogen level; WBC, white blood cell count; and NS: not significant. †*P* values were calculated using one-way factorial ANOVA.

**APTT, sec** Male 5 < 20 < 20 < 20 < 20 < 20

**/µl** Male 5 21.3 ± 0.4 31.6 ± 10.8 18.1 ± 4.4 25.0 ± 8.6 24.9 ± 5.1 NS

**/µL** Male 5 742.7 ± 72.6 858.0 ± 97.7 894.8 ± 55.8 919.0 ± 21.0 866.2 ± 24.5 < .05

Female 5 727.0 ± 20.2 886.6 ± 62.2 921.2 ± 64.5 901.4 ± 46.1 838.4 ± 44.2 < .001

Female 5 14.9 ± 0.4 17.5 ± 0.8 18.0 ± 0.9 18.4 ± 1.1 17.5 ± 0.6 < .001

Female 5 49.0 ± 1.8 56.1 ± 2.2 57.8 ± 4.2 57.9 ± 3.0 54.8 ± 2.8 < .01

Female 5 34.5 ± 2.0 24.8 ± 5.5 19.0 ± 5.0 24.8 ± 8.9 19.7 ± 5.7 < .05

Female 5 - 15.8 ± 1.1 16.1 ± 0.5 16.4 ± 0.5 16.0 ± 0.7 NS

Female 5 - 160.2 ± 19.4 145.2 ± 16.3 176.5 ± 20.1 123.3 ± 27.5 < .05

**Figure 1.** Number of genes expressed in whole blood of miniature pigs at different ages. In the graph, represents male and represents female. Values are means±SD. doi:10.1371/journal.pone.0019761.g001

#### **4.3. Microarray gene expression profiles – Correlation of gene expression**

Variations in correlation coefficients among individuals of the same age and different age groups were evaluated. Pearson correlation coefficient was used for correlation analysis. Correlation coefficients for a total of 31 microarrays were obtained in normalized signals log-scale after excluding "absent" spots. A color-coded pairwise correlation matrix is shown in Figure 2. The color scale at the bottom indicates correlation strength.

The average correlation coefficient within the same age group is shown in Figure 3. Variations in gene expression were greater for younger subjects, but it diminished with age while generating resembling expression patterns. Correlation coefficient within 30 weeks age group was slightly smaller than that within 20 weeks age group. However, this difference is smaller than other distant age groups. Significant differences were observed between any age groups according to an ANOVA analysis using Fisher's Z-transform. The average correlation coefficient between different age groups is shown in Figure 4. Significant differences were observed except between "fatal stage vs. 20 weeks" and "fatal stage vs. 30 weeks", and between "12 weeks vs. 20 weeks" and "12 weeks vs. 30 weeks" according to an ANOVA analysis using Fisher's Z-transform (P < 0.05). These results suggest that the variation in gene expression intensity within the same age was great in fetal stage and infancy period, but converged with age.

Whole Blood RNA Analysis, Aging and Disease 107

0.95 ± 0.03

**Figure 3. Age-related correlation coefficients within the same age groups.** Correlation coefficients were calculated between individuals within the same age groups. The bottom and top of the boxes represent the 25th and 75th percentiles respectively. The lower and upper whiskers denote the minimum and maximum values of the data. Comparisons of the groups were made with the ANOVA

0.87 ± 0.04

**Fetal stage 12 weeks 20 weeks 30 weeks**

0.90 ± 0.04

\* \*\* \*\* \*\* \*\* \*\*

0.98 ± 0.00

**Figure 4. Age-related correlation coefficients between the different age groups.** Correlation coefficients were calculated between the different age groups. The bottom and top of the boxes represent the 25th and 75th percentiles respectively. The lower and upper whiskers denote the minimum and maximum values of the data. Comparisons of the groups were made with the ANOVA

**Fetal stage**

**vs. 30 weeks**

**12 weeks**

**vs. 20 weeks**

**12 weeks**

**vs. 30 weeks**

**20 weeks**

**vs. 30 weeks**

**Fetal stage**

**vs. 20 weeks**

test. \* p < 0.05, \*\* p < 0.01.

0.7

0.4

**Fetal stage**

**vs. 12 weeks**

\*

\*

0.5

0.6

0.7

**Correlation coefficients**

0.8

0.9

1.0

0.8

0.9

**Correlation coefficients**

1.0

1.1

test. \* p < 0.01.

**Figure 2. Correlation matrix of age-related gene expression.** This color-coded correlation matrix illustrates pairwise correlations between the levels of gene expression in individuals. Probe sets with normalized signals (log-transformed and scaled) were used to calculate correlations between 31 arrays using Pearson correlation coefficient; signals flagged as "absent" were excluded. doi:10.1371/journal.pone.0019761.g002
