Few alterations in clinical pathology and histopathology observed in a CYP2C18&19 humanized mice model

  • Susanne Löfgren
  • Stina Ekman
  • Ylva Terelius
  • Ronny Fransson-Steen
Open Access
Research

Abstract

Background

This study was performed to characterize a gene-addition transgenic mouse containing a BAC (bacterial artificial chromosome) clone spanning the human CYP2C18&19 genes (tg-CYP2C18&19).

Methods

Hemizygous tg-CYP2C18&19, 11 week old mice were compared with wild-type littermates to obtain information regarding clinical status, clinical pathology and anatomical pathology. After one week of clinical observations, blood samples were collected, organs weighed, and tissues collected for histopathology.

Results

In males, the tissue weights were lower in tg-CYP2C18&19 than in wild-type mice for brain (p ≤ 0.05), adrenal glands (p ≤ 0.05) and brown fat deposits (p ≤ 0.001) while the heart weight was higher (p ≤ 0.001). In female tg-CYP2C18&19, the tissue weights were lower for brain (p ≤ 0.001) and spleen (p ≤ 0.001) compared to wild-type females. Male tg-CYP2C18&19 had increased blood glucose levels (p ≤ 0.01) while females had decreased blood triglyceride levels (p ≤ 0.01).

Conclusion

Despite the observed alterations, tg-CYP2C18&19 did not show any macroscopic or microscopic pathology at the examined age. Hence, these hemizygous transgenic mice were considered to be viable and healthy animals.

Background

The human cytochrome P450 enzymes from the 2C subfamily (CYP2C) are fairly well characterized and are known to metabolise many clinically important drugs. Four members belonging to the CYP2C family are found in man, namely CYP2C8, CYP2C9, CYP2C18 and CYP2C19 [1]. The anticancer drug paclitaxel is metabolised by CYP2C8 and the 6-hydroxylation of this compound is commonly used as a marker for this enzyme [2]. CYP2C9 metabolises many drugs, for example the hypoglycaemic drug tolbutamide [3], the anticonvulsant phenytoin [3, 4], the anticoagulant warfarin [5] and a number of nonsteroidal anti-inflammatory drugs including diclofenac and ibuprofen [6], which have all been used as marker substrates. The CYP2C18 protein has not yet been found in detectable amounts in any tissues [7], and its in vivo function is, to date, unknown. CYP2C19 stereo-selectively metabolises the S-enantiomer of the anticonvulsant mephenytoin to the metabolite 4-hydroxy- (S)-mephenytoin [8], and this metabolite is commonly measured to determine CYP2C19 activity in vitro. The 5-hydroxylation of R-omeprazole is selectively performed by CYP2C19 [9] and this reaction is also occasionally used as marker reaction for CYP2C19. A variety of other substrates are known to be metabolised by CYP2C19, such as the biguanide antimalarials [10], certain barbiturates [11], the β-blocker propranolol [12], the anxiolytic diazepam [13] and the antidepressant imipramine [14].

In contrast to the relatively small human CYP2C family, the mouse Cyp2c family is one of the largest and most complex, with 15 members published to date [1](for an update, see http://drnelson.utmem.edu/CytochromeP450.html). Cyp2c29 was the first mouse Cyp2c member identified [15], followed by Cyp2c37, Cyp2c38, Cyp2c39, Cyp2c40 [16], Cyp2c44 [17], Cyp2c50, Cyp2c54 and Cyp2c55 [18]. Six additional murine Cyp2c enzymes have thereafter been identified; Cyp2c65, Cyp2c66, Cyp2c67, Cyp2c68, Cyp2c69 and Cyp2c70 [1]. Their metabolic preferences are poorly characterized but their organ distribution is partially known [16, 17, 18, 19].

Murine CYP2C enzymes are involved in the metabolism of arachidonic acid, but the products formed differ between the isoforms. Human CYP2C19, on the other hand, is inhibited by the presence of arachidonic acid [20].

The transgenic mouse presented in this paper contains, in addition to all mouse Cyp2c enzymes, human CYP2C18 and CYP2C19. The inserted human CYP2C18 and CYP2C19 genes are expressed at the mRNA and protein levels, and the inserted CYP2C19 genes have been shown to be functional in vitro in metabolism studies using the CYP2C19 substrates S-mephenytoin and R-omeprazole)[21].

The aim of the present study was to characterize the humanized CYP2C18&19 mouse model as a basis for upcoming pharmacokinetic and toxicological studies. Hemizygous humanized CYP2C18&19 mice (tg-CYP2C18&19) were compared with wild-type littermates to obtain information regarding clinical status, body weight, clinical pathology, anatomy and morphology of this particular mouse model.

Methods

Generation of BAC transgenic mice

The transgenic CYP2C18&19 mice characterized in this article were generated as previously described)[21]. In brief, a BAC (bacterial artificial chromosome) clone named BAC RP11-466J14, which contains the CYP2C18 and CYP2C19 genes was purified. BAC DNA was injected into C57BL/6JOlaHsd (C57BL/6) eggs. Founders were identified by genotyping of DNA extracted from tail or ear biopsies.

Genotyping

For PCR detection of the inserted gene segment, gDNA was extracted from tail or ear biopsies either by using established protocols [22] or commercially available kits (DNeasy® Tissues, Qiagen). The gDNA was amplified in a 20 μL reaction mixture containing 10 μL HiFi PCR MasterMix (ABgene House, Surrey, UK), primers (250 nM of each primer for males or, alternatively, 500 nM of each primer for females) and 1 μL of gDNA. The four different specific primer pairs used are listed in Table 1. Cycling conditions were 94°C for 2 minutes (denaturation) and then 30 cycles of 94°C for 10 seconds, 60°C for 20 seconds, and 68°C for 45 seconds, followed by a 3 minutes extension at 70°C. The amplification products were analyzed on 1% agarose gels and the amplicons visualized with ultraviolet light.
Table 1

Sequences of primers used for genotyping of the mice (to detect the 466J14 BAC clone containing human CYP2C18 and CYP2C19)

Primer

Primer sequence (5'-3')

PCR product length (bp)

BAC5'endF

TAACATTAGCAGGTGAAGCCCAAA

706

BAC5'endR

CAATCTGTTCCATGATGGTTGATG

 

BAC3'endF

AGACTGTGCTATCATGGGAACCAA

480

BAC3'endR

GTTTTCTTGGGCTGAATGTCCTCT

 

2C18intron6F

GGCAAGAAACACTTCATGAGCACT

429

2C18intron6R

ATTCAGTTAAGGCCTCCCTTTTCC

 

2C19intron5F

CAAGATGGGCCTTATAAAGTTGGC

727

2C19intron5R

GAAGAAATTGGAACCCTCATGTCC

 

Animal husbandry

The hemizygous tg-CYP2C18&19 and the C57BL/6JOlaHsd (C57BL/6) wild-type littermates used were generated by crossing hemizygous tg-CYP2C18&19 males with C57BL/6JOlaHsd (C57BL/6) wild-type female mice. Wild-type littermates were used as controls. Attempts were also performed to generate homozygous mice by crossing hemizygous tg-CYP2C18&19 males and females, but the offspring died within a few days from birth.

For logistic reasons, the male and female groups were housed at different sites, but all mice were kept under conventional conditions and had free access to standard rodent diet (Males: R&M 1.E. SQC, pelleted diet, supplied by Special Diets Services Ltd, England; Females: RM3 Extended Breeding, supplied by Special Diets Services Ltd, England) and tap water. The animal husbandry and experimental conditions were approved by the Swedish Animal Welfare Agency.

Analysis of weight gain and food consumption

All mice were observed for one week prior to necropsy at an age of approximately 11 weeks. During that week, any adverse clinical signs observed on ocular inspection were recorded and the body weight gain and food consumption were measured. Individual body weights were recorded four and six days before necropsy for males and three and six days before necropsy for females. Six males and six females from each genotype (wild-type and tg-CYP2C18&19) were examined.

Pathological evaluation, total body weight and tissue weights

The total body weight of each mouse was determined, to the nearest 0.1 g, prior to necropsy and animals were killed by exsanguination of the common carotid artery under enflurane and nitric oxide anaesthesia. During necropsy, the organs were examined macroscopically and weights of a standard set of tissues (Table 2) were measured, to the nearest mg, prior to fixation. For bilateral organs, the total weight of the pair was recorded.
Table 2

Tissues sampled at necropsy

Tissue

Weighed at necropsy

Tissue

Weighed at necropsy

Adrenal glands

Yes

Muscle-skeletal

 

Aorta (thoracic)

 

Nerve-sciatic

 

Bone and bone marrow (sternum)

 

Optic nervesa

 

Brain

Yes

Ovaries

Yes

Brown fat deposita

Yes

Pancreas

 

Cervix

 

Parathyroid glandsb

 

Epididymides

 

Pituitary gland

 

Epididymal fat deposit

Yes

Prostate gland-ventral

Yes

Oesophagusa

 

Retriperitoneal fat deposit

Yes

Eyes

 

Salivary gland-parotid

 

Femur/femoro-tibial jointa

 

Salivary gland-submaxillary/lingual

 

Harderian glanda

 

Seminal vesicles

 

Heart

Yes

Skin

 

Intestine-duodenum

 

Spleena

Yes

Intestine-jejunum

 

Spinal cord-lumbar and cervical

 

Intestine-ileum

 

Stomach

 

Intestine-colon

 

Testes

Yes

Intestine-caecum

 

Thymus

Yes

Intestine-rectum

 

Thyroid glandsb

 

Kidneys

Yes

Tongue

 

Liver with gallbladder

Yes

Trachea

 

Lungs

Yes

Urinary bladder

 

Lymph node-mandibular

 

Uterus

Yes

Lymph node-mesenterica

 

Vagina

 

a/Tissue lost during processing in 1–2 animals and was not evaluated histologically.

b/Tissue lost during processing in all females but one per genotype. The tissue was present on slides from all male mice.

Forty-eight tissues (Table 2) from each mouse were collected and fixed. Eyes were fixed in MFAA (Methanol, Formalin, Acetic Acid); testicles and epididymides were fixed in Bouin's solution and all other tissues in 4% buffered formaldehyde. All tissues preserved were dehydrated, embedded in paraffin and cut into 4 μm sections before they were stained with haematoxylin and eosin for microscopic evaluation.

Clinical pathology parameters and analytical methods

Blood samples for haematology (EDTA tubes) and blood chemistry (lithium heparin tubes) were collected from the orbital plexus under enflurane and nitric oxide anaesthesia, prior to necropsy. Animals were not fasted at the time of blood sampling but the genotype groups were necropsied with every second animal being wild-type and every second being transgenic in order to minimize daytime variations in glycogen content between the groups. At necropsy, one femur was taken for bone marrow analysis. Haematology analysis was performed with an ADVIA® 120 Haematology System (Bayer Corporation, Diagnostic Division, Tarrytown, US) using standard methodology. Blood chemistry parameters were analysed using a Cobas Integra 400 analyser (Roche Diagnostics Instrument Centre, Switzerland), and appropriate kits. Parameters measured in haematology and blood chemistry are listed in Table 3.
Table 3

Parameters measured in clinical pathology

Haematology

Blood chemistry

Basophils (Baso)

Albumin (Alb)

Erythrocytes (RBC)

Albumin/globulin ratio (A/G)

Eosinophils (Eosn)

Alkaline aminotransferase (ALT)

Hematocrit (Hct)

Alkaline phosphatase (ALP)

Haemoglobin (Hgb)

Aspartate aminotransferase (AST)

Large unstained cells (LUC)

Bilirubin (total) (Bil)

Leucocytes (WBC)

Calcium (Ca)

Lymphocytes (Lymp)

Cholesterol (Chol)

Mean corpuscular haemoglobin (MCH)

Creatinine (Cre)

Mean corpuscular haemoglobin concentration (MCHC)

Glucose (Glu)

Mean red cell volume (MCV)

Potassium (K)

Monocytes (Mono)

Sodium (Na)

Neutrophils (Neut)

Total protein (TP)

Platelets (Plt)

Triglycerides (TG)

Red cell distribution width (RDW)

Urea (Urea)

Reticulocytes (Retc)

 

The bone marrow differentials were determined by flow cytometry as described by Saad et al [23] and the total nucleated cell count, the myeloid: erythroid ratio and the proportion of lymphoid, myeloid, erythroid and nucleated cells were determined. In addition, the proportion of cells staining positive for LDS-751 (laser dye styryl-751) was determined.

Statistical analysis

The statistical comparisons between the transgenic and wild-type groups were performed, using a computerized statistical program (Sigma Stat, version 2.03). All variables were compared with a t-test. Organ weights were compared both as absolute weights and as relative weights (relative to brain and relative to total body weight, data not shown). Box plots in figures were generated using Sigma Plot, 2001 and the plots show the median (line within each box), the 25th/75th percentile (outer boundaries of each box) and the 10th/90th percentiles (whiskers above and below each box). In addition the minimum and maximum are marked with dots.

Results

Weight gain and food consumption

All male mice either retained or increased their weight during the in vivo part of the study. In the female groups most of the mice gained weight. Two wild-type females and two tg-CYP2C18&19 females decreased in weight, but the weight loss in all cases was ≤ 0.6 g. No significant differences in food consumption were recorded.

Pathological evaluation

At necropsy, all mice were in good nutritional condition. On macroscopical examination, small white foci were found in the eyes of one wild-type female and two tg-CYP2C18&19 females. These macroscopical changes, and the recorded weight differences between groups, did not correlate to any changes on the histopathological level. Minimal perivascular infiltration of neutrophils was found on microscopic examination in the epididymal fat in one tg-CYP2C18&19 male and minimal alveolar histiocytosis was present in one tg-CYP2C18&19 female. All these changes were considered to belong to the spontaneous background pathology observed in laboratory mice of the C57BL/6JOlaHsd strain.

Total body weight and tissue weights

Weight distributions for all tissues within the different genotype and sex groups are shown in Table 4 and the statistically significant organ weight alterations are shown in Figure 1. Organ weights were compared between genotypes, both as absolute weights, relative to brain weight and relative to the total body weight. The relative data are not presented, since the same organs showed statistically significant weight differences between genotypes regardless of which comparison was used. The only exception was the brain weight relative to total body weight in male mice, which did not show any statistical difference between the transgenic and the wild-type groups.
Figure 1

Comparison of tissue weights between CYP2C18&19 transgenic and wild-type mice. The figure shows the absolute tissue weights for tissues with statistically significant differences between genotypes. Each group contains 6 animals. Asterisks indicate significant differences between groups, *p ≤ 0.05, ***p ≤ 0.001. wt: wild-type C57BL/6 mice, tg: hemizygous transgenic mice containing human CYP2C18&19. The box plots are presented as described in materials and methods.

Table 4

A comparison of absolute tissue weights between CYP2C18&19 transgenic and wild-type mice

 

Tissue

Wt

Tg

Wt

Tg

Body (g)

22.3 ± 1.5

22.2 ± 1.3

21.4 ± 0.5

21.0 ± 0.6

Brain (mg)

446 ± 11.4

427 ± 16.6*

472 ± 9.93

444 ± 11.9***

Heart (mg)

130 ± 7.39

149 ± 8.06***

139 ± 6.80

139 ± 6.05

Lung (mg)

148 ± 15.4

145 ± 5.61

159 ± 12.8

157 ± 18.9

Liver (mg)

1286 ± 76.2

1267 ± 102

1325 ± 89.7

1271 ± 54.2

Kidney (mg)

291 ± 18.9

286 ± 18.8

328 ± 9.51

315 ± 12.9

Adrenal glands (mg)

9.83 ± 3.25

5.83 ± 1.47*

9.33 ± 2.42

7.67 ± 2.16

Spleen (mg)

64.2 ± 7.41

61.0 ± 10.6

88.8 ± 3.49

75.0 ± 6.75***

Thymus (mg)

41.2 ± 7.31

37.7 ± 13.2

60.5 ± 7.45

68.0 ± 11.1

Retriperitoneal fat (mg)

69.7 ± 25.9

61.5 ± 8.07

25.8 ± 11.5

26.3 ± 7.61

Brown fat (mg)

95.0 ± 11.0

48.2 ± 17.9***

66.3 ± 13.5

60.2 ± 12.3

Testes (mg)

209 ± 19.3

210 ± 12.1

  

Prostate (mg)

39.3 ± 12.5

49.2 ± 9.11

  

Epididymal fat (mg)

306 ± 80.0

287 ± 37.0

  

Uterus (mg)

  

89.5 ± 43.0

112 ± 47.0

Ovaries (mg)

  

10.8 ± 6.68

13.3 ± 6.35

The table shows the average absolute tissue weights for each group +/- standard deviations. The statistical comparisons were performed as described in materials and methods. *p ≤ 0.05, ***p ≤ 0.001. wt: wild-type C57BL/6 mice, Tg: hemizygous transgenic mice containing human CYP2C18&19.

In both male and female mice, the brain weight was lower in tg-CYP2C18&19 than in wild-type mice (p ≤ 0.05 and p ≤ 0.001 for males and females respectively). The adrenal glands (p ≤ 0.05) and brown fat deposits (p ≤ 0.001) were smaller, while the heart weight was larger (p ≤ 0.001) in the tg-CYP2C18&19 males than in wild-type males. The spleen weight was lower in female tg-CYP2C18&19 than in wild-typefemales (p ≤ 0.05). All other organ weight comparisons (lung, liver, kidney, thymus, retriperitoneal fat deposits, testis, prostate, epididymal fat deposits, uterus and ovaries) showed no significant differences between the genotypes.

Clinical pathology parameters

Comparisons of all clinical pathology variables (between the genetic and sex groups) are shown in Table 5. Clinical pathology parameters with significant differences (p ≤ 0.05) between the genotypes are shown in Figure 2.
Figure 2

Comparisons of clinical pathology parameters between CYP2C18&19 transgenic and wild-type mice. The figure shows the blood chemistry parameters with statistically significant differences between genotypes. Each group contains 6 animals. Asterisks indicate significant differences between groups, **p ≤ 0.01. wt: wild-type C57BL/6 mice, tg: hemizygous transgenic mice containing human CYP2C18&19. The box plots are presented as described in materials and methods.

Table 5

Comparisons of haematology and blood chemistry parameters between genotypes of both sexes

Haematology parameter

Blood chemistry parameter

 

Wt

Tg

Wt

Tg

 

Wt

Tg

Wt

Tg

RBC (*1012/L)

8.89 ± 0.213

8.72 ± 0.200

9.78 ± 0.496

9.54 ± 0.286

GLU (mmol/L)

12.2 ± 0.708

13.9 ± 0.88**

15.3 ± 0.97

14.0 ± 1.33

HGB (g/L)

136 ± 1.51

134 ± 1.97

148 ± 8.80

146 ± 4.90

UREA (mmol/L)

7.10 ± 1.58

6.62 ± 1.20

8.58 ± 1.16

8.03 ± 1.33

Hct (%)

43.0 ± 0.894

42.3 ± 0.816

48.0 ± 1.79

47.3 ± 1.21

TP (g/L)

49.2 ± 1.47

48.2 ± 0.75

46.3 ± 1.21

45.7 ± 1.21

MCV (fL)

48.5 ± 1.18

48.7 ± 0.853

49.2 ± 0.809

49.6 ± 0.501

ALB (g/L)

28.0 ± 1.27

27.5 ± 0.548

29.8 ± 0.983

29.5 ± 1.23

MCH (pg)

15.3 ± 0.273

15.4 ± 0.303

15.1 ± 0.293

15.3 ± 0.341

A/G ratio

1.33 ± 0.121

1.33 ± 0.0516

1.80 ± 0.0632

1.85 ± 0.176

MCHC (g/L)

316 ± 4.50

316 ± 4.41

307 ± 8.31

308 ± 6.05

Na (mmol/L)

148 ± 1.03

147 ± 1.38

146 ± 0.516

147 ± 0.837

PLT (*1012/L)

1.18 ± 0.0394

1.18 ± 0.0299

1.04 ± 0.0714

1.07 ± 0.0659

K (mmol/L)

4.42 ± 0.475

4.32 ± 0.313

3.90 ± 0.329

3.83 ± 0.344

WBC (*109/L)

5.21 ± 1.95

4.37 ± 1.05

2.84 ± 1.28

2.32 ± 0.792

ALT (U/L)

79.2 ± 25.4

63.2 ± 18.9

60.3 ± 18.3

45.0 ± 3.35

Neutr (*106/L)

823 ± 752

470 ± 220

597 ± 646

483 ± 402

AST (U/L)

108 ± 62.1

77.0 ± 18.9

110 ± 58.5

93.5 ± 34.2

Lymp (*109/L)

3.99 ± 1.46

3.65 ± 0.975

2.04 ± 0.972

1.71 ± 0.589

ALP (U/L)

111 ± 8.98

113 ± 11.3

122 ± 13.6

134 ± 16.2

Mono (*106/L)

260 ± 237

86.7 ± 24.2

137 ± 95.8

66.7 ± 30.1

TBil (μmol/L)

7.66 ± 1.97

6.83 ± 1.33

11.7 ± 3.33

10.2 ± 1.72

Eosn (*106/L)

107 ± 35.0

127 ± 62.8

63.3 ± 55.7

46.7 ± 35.0

Crea (μmol/L)

8.33 ± 1.86

8.50 ± 1.52

7.83 ± 1.47

7.00 ± 0.894

Baso (*106/L)

6.67 ± 10.3

3.33 ± 8.16

10.0 ± 16.7

0 ± 0

Ca (mmol/L)

2.50 ± 0.0565

2.50 ± 0.0446

2.37 ± 0.0681

2.34 ± 0.0186

LUC (*106/L)

30.0 ± 21.0

30.0 ± 11.0

6.67 ± 10.3

0 ± 0

Chol (mmol/L)

2.42 ± 0.117

2.43 ± 0.163

1.70 ± 0.179

1.67 ± 0.197

Reti (*109/L)

283 ± 27.5

288 ± 15.1

271 ± 36.7

295 ± 30.7

TG (μmol/L)

795 ± 327

848 ± 372

555 ± 50.9

408 ± 83.3**

RDV (%)

12.6 ± 0.147

12.6 ± 0.261

12.9 ± 0.388

12.6 ± 0.210

     

The table shows the average haematology and blood chemistry values for each group +/- standard deviations. The statistical comparisons were performed as described in materials and methods, and the abbreviations used are explained in Table 3. **p ≤ 0.01.

wt: wild-type C57BL/6 mice, Tg: hemizygous transgenic mice containing human CYP2C18&19.

The blood glucose levels were altered in male tg-CYP2C18&19 mice, which had higher levels than wild-type males (p ≤ 0.01). The level of circulating triglycerides in the blood was decreased in female tg-CYP2C18&19 compared to the wild-type females (p ≤ 0.01).

All other haematology parameters and blood chemistry parameters measured showed no significant differences between the genotypes (See Table 3). There were no significant differences between the genotypes in any of the bone marrow parameters measured (total nucleated cell count, myeloid: erythroid ratio and the proportion of lymphoid, myeloid, erythroid and nucleated cells). The bone marrow parameter results are shown in Table 6.
Table 6

Comparisons of bone marrow parameters between genotypes of both sexes

Bone marrow parameter

 

Wt

Tg

Wt

Tg

TNC (*106/femur)

8.78 ± 0.995

9.27 ± 0.726

14.4 ± 2.71

11.7 ± 1.35

% Erythroid

37.8 ± 2.89

38.2 ± 3.51

33.6 ± 3.76

34.8 ± 5.88

% Lymphoid

12.0 ± 1.55

11.0 ± 1.40

8.41 ± 1.70

10.2 ± 1.48

% Myeloid

50.1 ± 3.38

50.6 ± 3.91

57.7 ± 4.81

54.6 ± 5.55

Ratio M: E

1.34 ± 0.188

1.35 ± 0.259

1.75 ± 0.342

1.64 ± 0.500

% LDS+

85.2 ± 2.19

85.7 ± 3.01

90.1 ± 2.60

89.4 ± 1.54

The table shows the average bone marrow values for each group +/- standard deviations. The statistical comparisons were performed as described in materials and methods.

wt: wild-type C57BL/6 mice, Tg: hemizygous transgenic mice containing human CYP2C18&19. TNC: Total Nucleated Count, % Erythroid/Lymphoid/Myeloid: Proportion (of total)

Erythroid/Lymphoid/Myeloid cells, Ration M: E: Myeloid: Erythroid ratio, %LDS+: percentage of (total) cells staining positive for the nucleic acid stain LDC-751 (laser dye styryl-751).

Discussion

The regulation and functions of the members of the CYP2C family are complex with large variations in expression within, and between, different tissues. The humanized mouse model characterized in this paper may facilitate the study and understanding of the functions of the human CYP2C enzymes. The demonstrated alterations in organ weights and clinical chemistry parameters cannot all be explained with the current knowledge of the CYP2C enzymes.

Arachidonic acid is metabolised in human brain parenchymal tissue to epoxyeicosatrienoic acid, which acts as a potent dilator of cerebral vessels [24]. In rats, this metabolism is carried out by CYP2C11 in the brain [25] and it is therefore possible that other members of the CYP2C subfamily, together with members of other cytochrome P450s subfamilies present in astrocytes, also participate in the metabolism of arachidonic acid in other species. The mouse Cyp2c enzymes are also involved in the metabolism of arachidonic acid and the murine isoforms metabolize arachidonic acid to regio- and stereospecific products[18]. This metabolism could be altered by the insertion of human CYP2C18&19 genes in the mouse model presented and, thereby, influence cerebral blood flow and possibly also the brain weight. The activities of the CYP2C enzymes in the central nervous system have also been proposed to influence the action of neurotransmitters, such as dopamine, which utilize fatty acid metabolites as intracellular mediators [26].

Alterations in the metabolism of arachidonic acid could possibly also explain the increased heart weight in the male tg-CYP2C18&19 mice. The CYP2C enzymes expressed in the cardiovascular system play a crucial role in the modulation of vascular homeostasis [27]. CYP products such as epoxyeicosatrienoic acids and reactive oxygen species have been implicated in the regulation of intracellular signalling cascades and vascular cell proliferation [28]. Preliminary behavioural studies show an initial increase in locomotor activity for male tg-CYP2C18&19 mice compared to wild-type controls when the mice are put in activity boxes. The increased activity of the transgenic mice could possibly also contribute to the increased heart weight (data not shown).

When focusing on lipid and glucose metabolism, the interactions are even more complex. In the present study, male tg-CYP2C18&19 mice had decreased brown fat deposits compared to wild-type mice and female tg-CYP2C18&19 had decreased levels of circulating triglycerides. The glucose levels were increased in male tg-CYP2C18&19 mice compared to wild-type males. Exogenous glucose administration to rats has been shown to decrease CYP2C6 and the male specific CYP2C11 activity by altering hepatic lipids [29]. If a similar male specific regulation occurs in the tg-CYP2C18&19 mice, this could possibly explain the alterations in glucose levels, fat deposits and blood triglyceride levels observed in this study.

Despite the few organ weight and clinical chemistry alterations observed, the hemizygous tg-CYP2C18&19 mice are considered to be viable and healthy. The alterations observed are also unlikely to cause any decrease in lifespan of the strain since the few tg-CYP2C18&19 male mice kept as breeders have reached an age of 2–3 years (unpublished data).

Conclusion

In the present study a gene-addition transgenic mouse, containing a BAC spanning the human CYP2C18&19 genes, has been characterized. Some alterations in organ weight and clinical pathology parameters were observed. Despite the alterations, no pathological changes were observed macroscopically or histologically and these hemizygous tg-CYP2C18&19 mice were considered to be viable and healthy. Hopefully, this model could be used to investigate the roles of CYP2C18 and CYP2C19 in vivo and extrapolation of results obtained from studies with this model may be more predictive to humans than when using traditional animal models.

Notes

Acknowledgements

We would like to thank all technical staff at the transgenic centre, DMPK and Safety Assessment at AstraZeneca for helping us with animal husbandry and various analyses. A special thank you to Yin Hu and Anna Wallin, at DMPK, who helped us with the genotyping during a period of heavy workload.

Supplementary material

13028_2008_496_MOESM1_ESM.jpeg (74 kb)
Authors’ original file for figure 1
13028_2008_496_MOESM2_ESM.pdf (76 kb)
Authors’ original file for figure 2

References

  1. 1.
    Nelson DR, Zeldin DC, Hoffman SM, Maltais LJ, Wain HM, Nebert DW: Comparison of cytochrome P450 (CYP) genes from the mouse and human genomes, including nomenclature recommendations for genes, pseudogenes and alternative-splice variants. Pharmacogenetics. 2004, 1: 1-18. 10.1097/00008571-200401000-00001.CrossRefGoogle Scholar
  2. 2.
    Rahman A, Korzekwa KR, Grogan J, Gonzalez FJ, Harris JW: Selective biotransformation of taxol to 6α-hydroxytaxol by human cytochrome P450 2C8. Cancer Res. 1994, 54 (21): 5543-5546.PubMedGoogle Scholar
  3. 3.
    Veronese ME, Mackenzie PI, Doecke CJ, McManus ME, Miners JO, Birkett DJ: Tolbutamide and phenytoin hydroxylations by cDNA-expressed human liver cytochrome P4502C9. Biochemical & Biophysical Research Communications. 1991, 3: 1112-8. 10.1016/0006-291X(91)91680-B.CrossRefGoogle Scholar
  4. 4.
    Bajpai M, Roskos LK, Shen DD, Levy RH: Roles of cytochrome P4502C9 and cytochrome P4502C19 in the stereoselective metabolism of phenytoin to its major metabolite. Drug Metab Dispos. 1996, 24 (12): 1401-1403.PubMedGoogle Scholar
  5. 5.
    Rettie AE, Korzekwa KR, Kunze KL, Lawrence RF, Eddy AC, Aoyama T, Gelboin HV, Gonzalez FJ, Trager WF: Hydroxylation of warfarin by human cDNA-expressed cytochrome P-450: a role for P-4502C9 in the etiology of (S)-warfarin-drug interactions. Chemical Research in Toxicology. 1992, 1: 54-9. 10.1021/tx00025a009.CrossRefGoogle Scholar
  6. 6.
    Leemann TD, Transon C, Bonnabry P, Dayer P: A major role for cytochrome P450TB (CYP2C subfamily) in the actions of non-steroidal antiinflammatory drugs. Drugs Exp Clin Res. 1993, 19 (5): 189-195.PubMedGoogle Scholar
  7. 7.
    Goldstein JA: Clinical relevance of genetic polymorphisms in the human CYP2C subfamily. British Journal of Clinical Pharmacology. 2001, 4: 349-55. 10.1046/j.0306-5251.2001.01499.x.CrossRefGoogle Scholar
  8. 8.
    Goldstein JA, Faletto MB, Romkes-Sparks M, Sullivan T, Kitareewan S, Raucy JL, Lasker JM, Ghanayem BI: Evidence that CYP2C19 is the major (S)-mephenytoin 4'-hydroxylase in humans. Biochemistry. 1994, 7: 1743-52. 10.1021/bi00173a017.CrossRefGoogle Scholar
  9. 9.
    Ibeanu GC, Ghanayem BI, Linko P, Li L, Pederson LG, Goldstein JA: Identification of residues 99, 220, and 221 of human cytochrome P450 2C19 as key determinants of omeprazole activity. J Biol Chem. 1996, 271 (21): 12496-12501. 10.1074/jbc.271.21.12496.CrossRefPubMedGoogle Scholar
  10. 10.
    Wright JD, Helsby NA, Ward SA: The role of S-mephenytoin hydroxylase (CYP2C19) in the metabolism of the antimalarial biguanides. Br J Clin Pharmacol. 1995, 39 (4): 441-444.PubMedCentralCrossRefPubMedGoogle Scholar
  11. 11.
    Knodell RG, Dubey RK, Wilkinson GR, Guengerich FP: Oxidative metabolism of hexobarbital in human liver: relationship to polymorphic S-mephenytoin 4-hydroxylation. J Pharmacol Exp Ther. 1988, 245 (3): 845-849.PubMedGoogle Scholar
  12. 12.
    Ward SA, Walle T, Walle UK, Wilkinson GR, Branch RA: Propranolol's metabolism is determined by both mephenytoin and debrisoquin hydroxylase activities. Clin Pharmacol Ther. 1989, 45 (1): 72-79.CrossRefPubMedGoogle Scholar
  13. 13.
    Sohn DR, Kusaka M, Ishizaki T, Shin SG, Jang IJ, Shin JG, Chiba K: Incidence of S-mephenytoin hydroxylation deficiency in a Korean population and the interphenotypic differences in diazepam pharmacokinetics. Clin Pharmacol Ther. 1992, 52 (2): 160-169.CrossRefPubMedGoogle Scholar
  14. 14.
    Chiba K, Saitoh A, Koyama E, Tani M, Hayashi M, Ishizaki T: The role of S-mephenytoin 4'-hydroxylase in imipramine metabolism by human liver microsomes: a two-enzyme kinetic analysis of N-demethylation and 2-hydroxylation. Br J Clin Pharmacol. 1994, 37 (3): 237-242.PubMedCentralCrossRefPubMedGoogle Scholar
  15. 15.
    Matsunaga T, Watanabe K, Yamamoto I, Negishi M, Gonzalez FJ, Yoshimura H: cDNA cloning and sequence of CYP2C29 encoding P-450 MUT-2, a microsomal aldehyde oxygenase. Biochimica et Biophysica Acta. 1994, 2–3: 299-301. 10.1016/0005-2728(94)90237-2.CrossRefGoogle Scholar
  16. 16.
    Luo G, Zeldin DC, Blaisdell JA, Hodgson E, Goldstein JA: Cloning and expression of murine CYP2Cs and their ability to metabolize arachidonic acid. Archives of Biochemistry & Biophysics. 1998, 1: 45-57. 10.1006/abbi.1998.0806.CrossRefGoogle Scholar
  17. 17.
    DeLozier TC, Tsao CC, Coulter SJ, Foley J, Bradbury JA, Zeldin DC, Goldstein JA: CYP2C44, a new murine CYP2C that metabolizes arachidonic acid to unique stereospecific products. Journal of Pharmacology & Experimental Therapeutics. 2004, 3: 845-54. 10.1124/jpet.104.067819.CrossRefGoogle Scholar
  18. 18.
    Wang H, Zhao Y, Bradbury JA, Graves JP, Foley J, Blaisdell JA, Goldstein JA, Zeldin DC: Cloning, expression, and characterization of three new mouse cytochrome P450 enzymes and partial characterization of their fatty acid oxidation activities. Molecular Pharmacology. 2004, 5: 1148-58. 10.1124/mol.65.5.1148.CrossRefGoogle Scholar
  19. 19.
    Tsao CC, Coulter SJ, Chien A, Luo G, Clayton NP, Maronpot R, Goldstein JA, Zeldin DC: Identification and localization of five CYP2Cs in murine extrahepatic tissues and their metabolism of arachidonic acid to regio- and stereoselective products. J Pharmacol Exp Ther. 2001, 299 (1): 39-47.PubMedGoogle Scholar
  20. 20.
    Yao HT, Chang YW, Lan SJ, Chen CT, Hsu JT, Yeh TK: The inhibitory effect of polyunsaturated fatty acids on human CYP enzymes. Life Sciences. 2006, 26: 2432-40. 10.1016/j.lfs.2006.08.016.CrossRefGoogle Scholar
  21. 21.
    Löfgren S, Baldwin RM, Hiratsuka M, Lindqvist A, Carlberg A, Sim AC, Shülke MSM, Edenro A, Fransson-Steen R, Terelius Y, Ingelman-Sundberg M: Generation of mice transgenic for human CYP2C18 and CYP2C19: characterization of the sexually dimorphic gene and enzyme expression. Drug Metabolism and Disposition. 2008, 5: 955-962. 10.1124/dmd.107.019349.CrossRefGoogle Scholar
  22. 22.
    Laird PW, Zijderveld A, Linders K, Rudnicki MA, Jaenisch R, Berns A: Simplified mammalian DNA isolation procedure. Nucleic Acids Research. 1991, 15: 4293-10.1093/nar/19.15.4293.CrossRefGoogle Scholar
  23. 23.
    Saad A, Palm M, Widell S, Reiland S: Differential analysis of rat bone marrow by flow cytometry. Comparative Haematology International. 2000, 97-101. 10.1007/s005800070016.Google Scholar
  24. 24.
    Harder DR, Alkayed NJ, Lange AR, Gebremedhin D, Roman RJ: Functional hyperemia in the brain: hypothesis for astrocyte-derived vasodilator metabolites. Stroke. 1998, 29 (1): 229-234.CrossRefPubMedGoogle Scholar
  25. 25.
    Alkayed NJ, Narayanan J, Gebremedhin D, Medhora M, Roman RJ, Harder DR: Molecular characterization of an arachidonic acid epoxygenase in rat brain astrocytes. Stroke. 1996, 27 (5): 971-979.CrossRefPubMedGoogle Scholar
  26. 26.
    Warner M, Stromstedt M, Wyss A, Gustafsson JA: Regulation of cytochrome P450 in the central nervous system. Journal of Steroid Biochemistry & Molecular Biology. 1993, 1–6: 191-4. 10.1016/0960-0760(93)90074-7.CrossRefGoogle Scholar
  27. 27.
    Fisslthaler B, Michaelis UR, Randriamboavonjy V, Busse R, Fleming I: Cytochrome P450 epoxygenases and vascular tone: novel role for HMG-CoA reductase inhibitors in the regulation of CYP 2C expression. Biochim Biophys Acta. 2003, 1619 (3): 332-339.CrossRefPubMedGoogle Scholar
  28. 28.
    Fleming I: Cytochrome P450 2C is an EDHT synthase in coronary arteries. Trends in Cardiovascular Medicine. 2000, 4: 166-70. 10.1016/S1050-1738(00)00065-7.CrossRefGoogle Scholar
  29. 29.
    Stewart CC, Strother A: Glucose consumption by rats decreases cytochrome P450 enzyme activity by altering hepatic lipids. Life Sciences. 1999, 23: 2163-72. 10.1016/S0024-3205(99)00165-4.CrossRefGoogle Scholar

Copyright information

© Löfgren et al; licensee BioMed Central Ltd. 2008

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Authors and Affiliations

  • Susanne Löfgren
    • 1
  • Stina Ekman
    • 2
  • Ylva Terelius
    • 3
  • Ronny Fransson-Steen
    • 1
  1. 1.Safety Assessment SwedenAstraZeneca R&D SödertäljeSödertäljeSweden
  2. 2.Department of Biomedical Sciences and Veterinary Public HealthDivision of Pathology, Pharmacology & ToxicologyUppsalaSweden
  3. 3.Medivir ABHuddingeSweden

Personalised recommendations