A doxycycline-inducible, tissue-specific aromatase-expressing transgenic mouse

Transgenic Res (2012) 21:415–428 DOI 10.1007/s11248-011-9525-7

TECHNICAL REPORT

A doxycycline-inducible, tissue-specific aromatase-expressing transgenic mouse Jenny D. Y. Chow • John T. Price • Margaret M. Bills Evan R. Simpson • Wah Chin Boon

•

Received: 6 December 2010 / Accepted: 13 May 2011 / Published online: 26 May 2011 Ó Springer Science+Business Media B.V. 2011

Abstract Aromatase converts androgens to estrogens and it is expressed in gonads and non-reproductive tissues (e.g. brain and adipose tissues). As circulating levels of estrogens in males are low, we hypothesize that local estrogen production is important for the regulation of physiological functions (e.g. metabolism) and pathological development (e.g. breast and prostate cancers) by acting in a paracrine and/or intracrine manner. We generated a tissuespecific doxycycline-inducible, aromatase transgenic mouse to test this hypothesis. The transgene construct (pTetOAROM) consists of a full-length human aromatase cDNA (hAROM) and a luciferase gene under the control of a bi-directional tetracycline-responsive promoter (pTetO), which is regulated by transactivators (rtTA or tTA) and doxycycline. Our in vitro

J. D. Y. Chow  E. R. Simpson  W. C. Boon Prince Henry’s Institute, Clayton, VIC 3168, Australia J. D. Y. Chow  W. C. Boon Department of Anatomy and Developmental Biology, Monash University, Clayton, VIC 3800, Australia J. T. Price  M. M. Bills  E. R. Simpson Department Biochemistry and Molecular Biology, Monash University, Clayton, VIC 3800, Australia W. C. Boon (&) Florey Neuroscience Institutes, The University of Melbourne, Parkville, VIC 3010, Australia e-mail: [email protected]

studies using MBA-MB-231tet cells stably expressing rtTA, showed that doxycycline treatment induced transgene expression of hAROM transcripts by 17-fold (P = 0.01), aromatase activity by 26-fold, (P = 0.0008) and luciferase activity by 9.6-fold (P = 0.0006). Pronuclear microinjection of the transgene generated four pTetOAROM founder mice. A male founder was bred with a female mammary gland-specific rtTA mouse (MMTVrtTA) to produce MMTVrtTA-pTetOAROM double-transgenic mice. Upon doxycycline treatment via drinking water, human aromatase expression was detected by RTPCR, specifically in mammary glands, salivary glands and seminal vesicles of double-stransgenic mice. Luciferase expression and activity was detected in these tissues by in vivo bioluminescence imaging, in vitro luciferase assay and RT-PCR. In summary, we generated a transgenic mouse model that expresses the human aromatase transgene in a temporal- and spatial-specific manner, which will be a useful model to study the physiological importance of local estrogen production. Keywords Conditional transgenic  Aromatase  Luciferase  Doxycycline  Tet-ON  MMTV Abbreviations dox Doxycycline hAROM Human aromatase gene/cDNA luc Firefly luciferase cDNA MMTV Mouse mammary tumour virus (promoter)

123

416

pTetO rtTA

Transgenic Res (2012) 21:415–428

Tetracycline-responsive promoter Reverse transactivation

Introduction Aromatase is the key enzyme responsible converting androgens to estrogens; it is highly expressed in premenopausal ovarian granulosa cells and corpus luteum (McNatty et al. 1976). Research in recent decades has highlighted the role of estrogens in the male, and that extra-gonadal tissues (including the brain, adipose tissue, bone, skin and vascular endothelia) also express aromatase and actively convert circulating androgens into local estrogens. This has important intracrine or paracrine effects on homeostasis such as cardiovascular and bone health, metabolism and behavior. These local sites are the major sources of endogenous estrogen in men, whose circulating levels of estrogens are low, and in postmenopausal women, whose ovaries cease to produce estrogens (Simpson 2003). The absence of this local estrogen production in aromatase-deficient men has been associated with the development of metabolic defects such as central obesity, insulin resistance, dyslipidemia, hepatic steatosis (Maffei et al. 2004, 2007), and continuous bone growth (Morishima et al. 1995; Carani et al. 1997). All these observations highlight the importance of intracrine and paracrine estrogen actions not only for reproductive functions but also non-reproductive functions such as metabolism. The importance of local estrogen production is also reflected in the complexity of the multiple tissue-specific promoters of the human aromatase (CYP19A1) gene. Eleven tissue-specific promoters of the CYP19A1 gene have been identified to date (Demura et al. 2008), whereas four tissue-specific promoters found in the mouse Cyp19A1 gene (Chow et al. 2009); each of which regulates the expression of aromatase in a particular tissue (Bulun et al. 2003). For example, aromatase expression is predominantly regulated by promoter II (PII) in ovaries, by promoter I.4 in adipose tissue and bone (Mahendroo et al. 1993). Promoter switching has been associated with disorders such as breast cancer, where PII becomes the dominant driver of

123

aromatase expression instead of promoter I.4 (Harada et al. 1993; Chen et al. 2009), associated with elevated local estrogen production in cancerous breast tissues (Sasano et al. 1994). Therefore, we hypothesize that estrogens produced locally in many different tissues have essential roles in maintaining the homeostasis of many processes in the body, particularly in men and postmenopausal women, where circulating levels of estrogens are low. To dissect the role of local estrogens in different organs, an inducible mouse model that expresses aromatase in a tissue-specific manner is required. The doxycycline-inducible system has been used to direct transgene expression in a tissue-specific manner (Gunther et al. 2002; Zhang et al. 2003). The system drives transgene expression using a tetracycline-responsive promoter, which is activated by transactivator molecules in the presence (Tet-ON) or absence (Tet-OFF) of doxycycline (Gossen and Bujard 1992). Tissue-specific transgene expression can be achieved by using a tissuespecific promoter to direct transactivator (tTA for Tet-OFF; rtTA for Tet-ON) expression in a particular tissue. Doxycycline-inducible expression systems provide greater control over transgene expression spatially as well as temporally to avoid non-specific expression under un-induced conditions. Moreover, doxycycline has no cytotoxic effects to the liver (Labbe et al. 1991); it is less pleiotropic than other inducing agents such as hormones and metal ions which naturally play a role in mammalian physiology (Yarranton 1992). The advantage of this current model is that the tetresponsive promoter (pTetO) is bidirectional, which means it can drive transgenes cloned on either side of the pTetO simultaneously. This is favorable as the gene of interest and a reporter gene can be cloned into the same expression vector as two separate genes rather than into two separate constructs (Albanese et al. 2000), or as a fused transgene which may result in fusion proteins. The current study has produced a novel transgene construct that enabled the generation of a doxycycline-inducible, tissue-specific aromatase-expressing mouse model, which will enable researchers to study the physiological significance of estrogens locally produced at any tissue of interest.

Transgenic Res (2012) 21:415–428

Materials and methods The pTetOAROM transgene The pTRE-tight-Bi vector (Clontech, USA) contains a bidirectional promoter (pTRE-tight) with a multiple cloning site (MCS I and II) on either side of promoter. The pTetOAROM transgene was generated by subcloning a 1.5 kb human aromatase cDNA [hAROM; (Simpson et al. 1987)] into MCS II (BamHI and EcoRI), and a 1.6 kb firefly luciferase cDNA (luc; from Clontech pTRE-tight-BI-Luc control vector) into MCS I (BglII and XbaI; Fig. 1). pTetOAROM expression in vitro The purified pTetOAROM plasmid (Maxiprep, Promega, USA) was transfected into MDA-MB-231tet cells, a human breast cancer cell line stably transfected with the rtTA transactivator. Cells were cultured in complete DMEM media (10% fetal calf serum, 1% L-Glutamine and 0.1% Penicillin–Streptomycin) in 12-well plate, 37°C and 5% CO2. At 70–80% confluency, cells were transfected with plasmid DNA (Fugene transfection reagent, Roche, Germany) for 4 h and cultured in 1 ml of fresh complete media with or without 1 lg/ml doxycycline (Clontech, USA) for 24 h for RTPCR and luciferase activity (350 ng plasmid) analyses, or 60 h for aromatase activity assay (1 lg plasmid). Luciferase assay Luciferase assay was conducted on cell lysate or tissue homogenates according to manucfacturer’s instructions (Luciferase Assay System, Promega, USA), in a 96-well plate and quantified immediately at 540 nm wavelength absorbance (Wallac 1420 Victor Plate Reader; LabX, Canada).

417

aromatase activity to be determined by the isolation and quantification of 3HOH. Protein concentration of cell lysate was determined using the standard colorimetric BCA Protein Assay Kit (Thermo Scientific, USA) and quantified at 490 nm wavelength (Wallac 1420 Victor Plate Reader, LabX, Canada). pTetOAROM expression in vivo and in tissues Four pTetOAROM founder mice were generated by pronuclei microinjection (MouseWorks Service, Australia) of the BsrBI transgene fragment from pTetOAROM plasmid. MMTVrtTA-pTetOAROM double-transgenic mice were generated by crossbreeding a pTetOAROM founder male with the female mammary-specific MMTVrtTA transgenic mouse (Gunther et al. 2002). All mice were housed in specific pathogen-free conditions, with ad libitum access to water and food. All experimental procedures were approved by Monash Medical Centre (B) Animal Ethics Committee. Litters were genotyped by genomic PCR to detect the presence of hAROM, mouse aromatase exon 9 and luc transgene with primers, 1–4 f and r, and to detect the presence of MMTVrtTA gene with primers 5f and 5r (Table 1) and gel electrophoresis (2% agarose, 19 TBE). Transgene copy number determination Genomic DNA extracted from WT and transgenic founder mice were amplified by real-time PCR (Rotor-Gene 3000, Qiagen, Germany) using primers 3f and 3r (Table 1), which anneals to exon 9 of both the hArom and mouse Cyp19A1. PCR conditions: 95°C 10 min 1 cycle; 95°C 10 s, 55°C 10 s, 72°C 9 s, 45–50 cycles. Results were normalized to cyclophilin gene (primers 8f and 8r, Table 1). Luciferase activity imaging

Tritiated water release assay Aromatase activity was measured in transfected cells by the tritiated water (3HOH) release assay (Ackerman et al. 1981; Lephart and Simpson 1991). The aromatase assay utilizes a radioactive substrate [1b-3H] androstenedione (Perkin Elmer, USA) which was labeled with 3H at the b position of the A-ring of androstenedione. During aromatization, the 1b-3H becomes incorporated into water, thus allowing the

Eight week-old female and male WT, MTB, pTetOAROM and MTB-pTetOAROM, mice were treated with doxycycline (2 mg/ml, 5% sucrose) in drinking water or normal drinking water (doubletransgenic only) for 72 h ad libitum. Before scanning, mice were anaesthetized by isoflurane and injected i.p. with luciferin (150 mg/kg body weight; Clontech, USA). After 15 min, animals were placed ventral recumbent position in the IVIS 200 bioluminescence

123

418

Transgenic Res (2012) 21:415–428

A

BsrBI, + 4066bp

PvuI, + 3595bp

+ 1bp

BsrBI, + 2,262bp

MCS II h-AROM cDNA (1528bp)

MCS I Luciferase (1669bp)

P tight (450bp)

BsrBI SV40 polyA

BsrBI

C Uncut

XbaI / Bgl II

Uncut

B

EcoRV

Ladder

SV40 polyA

Ladder

BamHI

Bgl II BamHI 614bp

Xba I

BamHI / EcoRV

BsrBI

5976bp 5976bp 4448bp

2kb

2kb 1528bp 1kb

1669bp 1078kb

Fig. 1 The pTetOAROM transgene. a The original pTRETight-BI empty vector (Clontech) was 2,800 bp in length, and it contains a bi-directional promoter, Ptight, which consists of two CMV promoters placed under the control of a tetracyclineresponsive element, TREmod. The firefly luciferase gene (1,669 bp) was first subcloned into the vector at multiple cloning site (MCS) I using the restriction enzymes BamH1 and EcoRV. The full-length human cDNA (hAROM, 1,528 bp including kozak sequence) was subcloned into MCS II using Xba I and Bgl II. The entire plasmid (5,976 bp) was then linearized using BsrBI, which cuts at two specific sites to release the transgene fragment (4,172 bp) that was later purified and microinjected to generate transgenic founder mice. b To verify that the plasmid contains the transgenes

123

4172 bp

3229bp

1804 bp

1kb

hAROM and Luc, enzyme restriction was performed and the products were analysed by gel electrophoresis. The entire transgene plasmid, uncut is 5,976 bp in size. Restrictions with Xba I and Bgl II released the hAROM cDNA fragment (1,528 bp) from the remaining plasmid (4,448 bp). Restrictions with BamHI and EcoRV released the marker gene luciferase (1,669 bp), as well as bisecting the remaining DNA into two fragments (3,229 and 1,078 bp) since Bam HI can also cut hAROM at ?614. The identity of both genes was confirmed by sequence-analysis. c The transgene plasmid was linearized using BsrBI to produce the transgene fragment (4,172 bp) and purified from the remaining vector DNA (1, 804 bp) for injection into pronuclei of mouse zygotes to generate transgenic founder

Transgenic Res (2012) 21:415–428

419

Table 1 Oligo primer sequences (Sigma) for subcloning, genotyping and real-time PCR No.

Name

Function

Sequence (50 ? 30 )

Tm (°C)

Product size (bp)

1f

AF

Subcloning: adds restriction site (BglII) and kozak sequence 50 to hArom cDNA and amplifies with 1r; For genotyping for the presence of hArom with 3r

GGA AGA TCT GGA ACA CAA GAT GGT TTT GGA AAT GCT G

61.6

1,540

1r

AR

Subcloning: adds restriction site (XbaI) 30 to hArom cDNA and amplifies with 1f

GCTC TAG ACT AGT GTT CCA GAC ACC TGT C

2r

RT8

Genotyping with 1f

CAG GAA TCT GCC GTG GGA GA

64

217

10

Zhou et al. (2005)

3f

E9-F

Genotyping: presence of WT mouse aromatase (exon 9); transgene copy number determination (real-time PCR)

GTG ACA GAG ACA TAA AGA TCG

55

220

30; or 9 (realtime PCR)

Robertson et al. (1999)

3r

E9-R

4f

Luc-F

55

436

30

Current study

4r

Luc-R

5f

MTB-F

65

500

60

Gunther et al. (2002)

5r

MTB-R

6f

hAROMf

59

242

10

Yamamoto et al. (2002)

6r

hAROMr

7f

mE5-f

58

175

7

Current study

7r

mE6-r

8f

Cyc-f

60

180

7

Boon et al. (2005)

8r

Cyc-r

Extension time (s)

Reference

Current study

92

CCC TAA GCC CAA TGA ATT TAC Genotyping for the presence of the luciferase gene, used with 3f and 3r

AGA CTT CAA GCG GTC AAC TAT CGC CAA AAG CAC TCT GAT

Genotyping for the presence of MMTVrtTA gene, and for One-step RT-PCR

TGC CGC CAT TAT TAC GAC AAG C ACC GTA CTC GTC AAT TCC AAG GG

RT-PCR: to amplify human aromatase transcript

ACC CTT CTG CGT CGT GTC A TCT GTG GAA ATC CTG CGT CTT

RT-PCR: to amplify mouse aromatase transcript, crossing Cyp19 exons 5 and 6

GGA GTC CAT CAA GCA GCA TT GGC GTT AAA GTA ACC CTG GA

RT-PCR: to amplify mouse cyclophilin housekeeping gene

CTT GGG CCG CGT CTC CTT C TGC CGC CAG TGC CAT TAT

imaging chamber (Caliper Life Sciences, USA) and imaged for 5 min with the highest sensitivity (large/ medium binning). Luciferase activity was measured using the Living Image software (Caliper Life Sciences) as photon counts.

Tissues Mice were killed by CO2 asphyxiation. Brain, liver, omental fat, gastrocnemia muscle, gonads, mammary glands (4th pair or inguinal), salivary glands, seminal

123

420

vesicles and uterus were harvested. Tissues were embedded in O.C.T and frozen for histological examination or snapped frozen in liquid nitrogen for transcripts and luciferase activity analyses. All procedures were approved by the Institutes Animal Ethics Committee.

Transgenic Res (2012) 21:415–428

(GraphPad Prism 4, GraphPad Software, USA). Differences were deemed significant when P \ 0.05.

Results The pTetOAROM transgene

RNA extraction

RNA (1 lg) was incubated with 50 lg Random Primers at 72°C, 5 min, and then reverse-transcribed in a total reaction volume of 25 ll containing 5 ll 59 RT buffer, 2.5 ll 10 mM dNTPs, 40U RNasin, 20U AMV-Reverse Transcriptase (Promega, USA) and DNase-free water at 95°C 5 min, 45°C 45 min, and 72°C 5 min. Transcripts were amplified using LightCycler-FastStart DNA Master SYBR Green I (Roche, Germany), in 10 ll total reaction volume containing 1 ll RT-mix, 0.8 ll 25 mM MgCl2, 0.5 ll 10 lM PCR primer (Table 1, 5–7f and r), 1 ll Reaction Mix and sterile DNase-free water. See Table 1 for PCR conditions.

The pTetOAROM transgene construct (Fig. 1a) contains a full-length human aromatase cDNA and a marker gene firefly luciferase, which were cloned into the pTRE-tight-Bi vector which can direct the expression of two genes of interest simultaneously in response to doxycycline. It contains two CMV promoters (PminCMV), which were placed under the regulation of a bi-directional tetracycline-responsive element (TREmod). The human AROM cDNA was cloned in multiple cloning site II whereas the luciferase cDNA is cloned into multiple cloning site I (Fig. 1a). Gel electrophoresis of the transgene plasmid restricted with different enzymes confirmed that the correct construct (Fig. 1b, c) was generated based on the sizes of predicted fragments from the restriction map. Sequence-analysis also confirmed the identities of both inserts. After validation, the plasmid was restricted with BsrBI at ?2,262 and ?4,066 positions on either side of the transgene fragment (4,172 bp; including both terminal SV40 polyA-tails) to release the transgene from the vector (1,804 bp). This transgene fragment was purified and microinjected into mouse zygotes to generate pTetOAROM transgenic founder mice. Whole plasmid was used for in vitro expression studies.

One-step RT-PCR

In vitro pTetOAROM expression

Total RNA (0.5 lg) was reverse transcribed amplified in a 50 ll reaction using SuperScriptTM III RT/ PlatinumÒTaq Mix (Invitrogen, USA). Reverse transcription was performed at 60°C 30 min.; See Table 1 for PCR conditions. Amplicons were analyzed by gel electrophoresis and gel-purified (Qiagen, Germany) for sequence analysis.

To test the expression efficiency, the transgenecontaining plasmid was transfected into rtTAexpressing MDA-MB231tet cells, treated with doxycycline (dox) to turn on expression, assayed for human aromatase transcript level, luciferase and aromatase activities. rtTA-expressing MDA-MB231tet cells were transfected with one of the following: empty pTREtight-Bi vector (vector only), plasmid containing only the human aromatase cDNA (hAROM only, for RT-PCR only), or the complete plasmid containing both luciferase and human aromatase cDNA

Total RNA was isolated from transfected cells using Qiagen RNA Extraction Kit (Qiagen, Germany) or 10–100 mg frozen tissues using UltraspecTM RNA isolation system (Biotecx, USA) according to manufacturers’ instructions. All RNA samples were DNase-treated (Ambion, USA) and the quality was analysed by gel electrophoresis and RNA concentration was determined by UV absorbance at 260 nm wavelength (Nanodrop spectrophotometer, Thermo Scientific, USA). Real-time RT-PCR

Statistical analysis Results were expressed as mean ± SEM and analyzed using the non-parametric Mann–Whitney test

123

Transgenic Res (2012) 21:415–428

2

*

0 Vector only

hAROM+Luc Tet cells

hAROM+Luc in non-tet cells

-1

Transfectant

Human aromatase transcript level 20

-Dox

**

+Dox

15 10

*

5 0 Vector only

hAROM only

hAROM + Luc

Transfectant

Aromatase activity

Tritiated water realeased [Cpm (vector subtracted) / mg protein]

C

-Dox +Dox

1

Arbituary units (Normalized to cyclophilin)

B

b Fig. 2 In vitro transgene expression. rtTA-expressing MDA-

Luciferase activity

Absorbance units / ug protein

A

421

MB231 cells were transfected with the empty pTRE-tight-Bi vector (vector only), plasmid containing only the human aromatase gene (hAROM-only), or plasmid containing both the luc and human aromatase cDNA (hAROM ? Luc). The hAROM ? Luc construct was also transfected into cells not expressing rtTA (non-tet cells; for luciferase assay). Cells were then either treated or not treated with dox. a Luciferase activity was not induced in cells transfected with the empty vector. Cells transfected with the hAROM ? Luc had a 9.6-fold increase in luciferase activity after dox compared to untreated. Dox did not induce luciferase activity in non-tet cells transfected with hAROM ? Luc. Luciferase activity was normalized to the amount of protein (lg). b Human aromatase transcript was quantified in reverse-transcribed total cell RNA using real-time PCR. Endogenous human aromatase transcript was detected in the human derived cell line. No significant induction was detected in empty vector-transfected cells. Dox induced human aromatase transcription in hAROM onlytransfected cells by 90-folds, and in hAROM ? Luc-transfected cells by 17-folds compared no dox control. Transcript level was normalized to cyclophilin. c Aromatase activity was measured using the standard tritiated water assay as detailed in ‘‘Materials and methods’’. Cells transfected with the empty vector did not have an increase in aromatase activity beyond endogenous level after dox treatment, whereas cells transfected with hAROM only or hAROM ? Luc had a 15-fold and 26-fold induction in aromatase activity respectively after dox treatment. All results were expressed as mean ± SEM, n = 3; statistical significance was indicated **P \ 0.01, *P \ 0.05

500

-Dox 400

**

+Dox

**

300 200 100 0 Vector only

hAROM only

hAROM + Luc

Transfectant

(hAROM ? Luc). To test that expression was dependent on the presence of rtTA, MBA-MD231 cells that do not express rtTA were also transfected with hAROM ? Luc for luciferase activity analysis. Transfected cells not treated with dox did not exhibit above background level of luciferase activity (P = 0.2; open bars; Fig. 2a). After dox-treatment, cells transfected with hAROM ? Luc plasmid had increased luciferase activity by 9.6-times (P \ 0.05) respectively over untreated cells (Fig. 2a). Doxtreatment did not induce luciferase activity in cells transfected with the empty vector. Human aromatase transgene expression was quantified by real-time PCR (Fig. 2b) and tritiated water

assay (Fig. 2c). There was endogenous hAROM transcript in the transfected, human-derived MDAMB231tet cell line. Cells transfected with hAROM or hAROM ? Luc transgenes had a 90- (P \ 0.001) and 17-fold (P \ 0.05) induction of aromatase transcription, respectively, compared to no dox-treatment (Fig. 2b). Endogenous aromatase activity was detectable by tritiated water assay in all uninduced transfected cells as well as induced vector only-transfected cells (Fig. 2c). Aromatase activity in cells transfected with hAROM or hAROM ? Luc constructs was increased by 15- (P \ 0.001) and 26-folds (P \ 0.001) after dox-treatment. pTetOAROM mice Four founder mice were successfully generated: female 448-1, male 449-3, female 452-1 and male 452-2. Founder 2 (449-3 male) had 4 copies of the pTetOAROM transgene inserted into the genome; the other three founders had 3 copies.

123

422

Transgenic Res (2012) 21:415–428

B

A MMTV-rtTA / pTetO-AROM -/-

pTetO-AROM

-

+

+

MMTV-rtTA

-

-

+

Dox

+

+

-/+ +/+ +/-

rtTA

+

-

Luc E9 h-Arom

Fig. 3 Genotyping and in vivo luciferase activity of MMTVrtTA-pTetOAROM mice. a The genotyping of MMTVrtTA-pTetOAROM mice includes triple PCRs to determine the presence of MMTVrtTA, Luc/E9 and hAROM. b In vivo luciferase activity assay. Eight week-old MMTVrtTApTetOAROM female and male wildtype (-/-), mono- (-/?) and double- (?/?) transgenic mice, after treatment with doxycycline-added drinking water or normal drinking water

(?/? only) for 72 h ad libitum as described in ‘‘Materials and methods’’. Representive scans were shown here, where luciferase activity was not detected in WT and -/? mice despite dox treatment. Male ?/? mice treated with dox had localized luciferase activity detected around the neck region, which was not seen in the same ?/? mouse untreated. This was not seen in female ?/? animals

Founders were bred with C57B6 WT mice to establish four separate lines of pTetOAROM transgenic mice (Lines 1–4, respectively). Founder mice presented normal phenotypes as the transgene is not active without the presence of transactivator expression. Due to the low number of pups, we were unable to maintain Line 1 to generation F2. Nonetheless, the other three lines were propagated normally to generation F2 to date.

(?/-) or no rtTA but had transgene (-/?). The absence of both was conferred WT (-/-) (Fig. 3a). Line 2 crosses produced only one litter of ten pups, none of which were double transgenic. Line 4 crosses produced four litters of 6–8 pups per litter: ?/? (18%), ?/- (36%), -/? (7%) and -/- (39%). To obtain higher frequencies of double-transgenics, three male ?/? mice from line 4 with the greatest in vivo luciferase activity detected were bred with six female MMTVrtTA mice to produce F2 double transgenic mice. The breeding produced eight litters of 4–11 pups per litter, and genotype distributions of ?/? (33%), ?/- (41%), -/? (10%) and -/- (16%). Mice from both generations were subjected to in vivo luciferase scanning and transgene expression analysis. In vivo bioluminescence imaging was performed to detect in vivo luciferase activity. In vivo luciferase activity was only observed at the neck and lower abdomen region (ventral) for dox-treated male ?/?

MMTVrtTA-pTetOAROM mice Founder mice of lines 2 and 4 were bred with female MMTVrtTA transgenic mice to generate MMTVrtTA-pTetOAROM double-transgenic mice. Positive (?/?) double-transgenic genotype will present all following PCR products: rtTA (500 bp), transgene (hArom, 214 bp; luciferase, 417 bp). Monotransgenic mice have rtTA with no transgene

123

Transgenic Res (2012) 21:415–428

423

assay also showed variable expression, where luciferase activity in the salivary gland of dox-treated male ?/? b59 was not any higher than that in b57 or b58 despite a more intense expression detected by the in vivo bioluminescence imaging (Fig. 4a). RT-PCR of human aromatase, mouse aromatase and luciferase transcripts was performed on the salivary gland (where detectable luciferase activity was most pronounced) of ?/? mice. High level of luciferase transcript was detected in the salivary gland of two dox-treated male ?/? mice, while the other two ?/? treated had the same low levels as

mice, and at the neck (ventral) of dox-treated female ?/? mice (Fig. 3b). Above background levels luciferase activity was not detected in dox-treated -/-, ?/-, -/? mice or in untreated ?/? mice (Fig. 3b). Tissues with high background levels were paws, teeth and the tail. Mice with black fur coat tended to have lower levels of luminescence due to the absorption of emitted photons by the dark coloured fur. The pattern and strength of luciferase expression is highly variable among male ?/? mice (Fig. 4a). This variability may be a result of inheriting single or double rtTA copies from parents. In vitro luciferase

Male tissue luciferase activity (individual)

Mammary

3.00

Seminal vesicle 2.00

1.00

0.00

+/+ dox

+/+ dox

+/+ dox

b57

b58

b59

Luciferase transcripts (Salivary gland) 0.008 0.006 0.004 0.002 0 +/+ dox

+/+ dox

+/+ dox

+/+ dox

+/+ no dox

-/+ -/- dox dox

+/+ dox

+/+ dox

+/+ dox

+/+ -/- dox No dox

71

97

98

100

67

96

61

72

94

91

87

male SG

Arbituary units (normalized to cyclophilin)

Absorbance units / protein mg/ml (background subtracted)

Salivary

Arbituary units (normalized to cyclophilin)

B

A

95

female SG

Human aromatase transcripts (Salivary gland)

0.025 0.02 0.015 0.01 0.005 0 +/+ dox

+/+ dox

+/+ dox

+/+ dox

71

97

98

100 male SG

Fig. 4 Variation in transgene expression. a Male MMTVrtTApTetOAROM (6 month-old) treated with dox as described in ‘‘Materials and methods’’, showed differential luciferase expression at salivary gland or seminal vesicle, as detected by both tissue luciferase assay (upper) and in vivo

+/+ no -/+ dox -/- dox dox 67

96

87

+/+ dox

+/+ dox

+/+ dox

61

72

94

+/+ No -/- dox dox 91

95

female SG

bioluminescence imaging. b Luciferase transcript and c human aromatase transcripts were amplified by RT-PCR from RNA of male and female salivary glands (dox-treated or not treated), and normalized to housekeeping cyclophilin and presented individually

123

424

Transgenic Res (2012) 21:415–428

Fig. 5 Transactivator rtTA transcript in mammary gland of MMTVrtTApTetOAROM mice. Transactivator (rtTA) and housekeeping cyclophilin (cyc) transcripts were amplified using one-step RT-PCR from male (M) and female (F) mammary glands (MMTV-rtTA and pTetO-AROM double transgenic or monotransgenic, with or without dox treatment). Salivary gland (SG) of an induced double-transgenic male mouse was used as a positive control, and water (W) as negative control. rtTA transcript was detectable only in male double transgenic doxinduced or not, but not in female double transgenic mice nor monotransgenic male mice

rtTA (500bp)

Cyc (180bp)

Mammary gland

untreated ?/? and -/-mice (Fig. 4b). All female mice had low levels of luciferase transcript in the salivary gland. Human aromatase transcript levels closely mirrored those of luciferase transcript (Fig. 4c), and mouse aromatase transcript was not detected in all salivary gland samples (data not shown). To investigate whether the lack of transgene expression was due to the absence of transactivator (rtTA) expression, one-step RT-PCR of rtTA was performed on mammary gland RNA samples of selected male and female transgenic mice (Fig. 5). After 40 PCR cycles, rtTA transcript was not detectable in female double transgenic mammary gland with or without dox treatment. It was not detectable, as expected, in -/? male mice. The rtTA transcript was clearly detectable in male ?/? mammary gland with or without dox treatment. However, for one male ?/?, rtTA transcript was only detectable in salivary gland but not in mammary gland (Fig. 5).

123

Salivary Negative gland

Discussion We have generated the first doxycycline-inducible mouse model that expresses aromatase in a tissuespecific manner. The bidirectional pTetO promoter allows the co-expression of a reporter gene (e.g. luciferase) with the gene of interest (e.g. aromatase), which enabled easier detection of transgene expression. Many previous studies employed separate expression vectors (Albanese et al. 2000) or fusion trangenes to co-express a gene of interest and a reporter gene, which may complicate the process of transgenesis or create fusion transgenic proteins that may fold inappropriately or non-physiologically. The previous use of a dox-inducible ‘super module vector’ was shown to direct the simultaneous expression of c-Src and two reporter genes. However such a large construct may pose a challenge in generating transgenic mice via microinjection, as the initial microinjection using a nonlinearized pTetOAROM

Transgenic Res (2012) 21:415–428

plasmid in this study did not result in any positive transgenic founder mice, possibly due to the considerable size of the entire plasmid (data not shown). Transgenic offspring of pTetOAROM appeared normal, with no notably different phenotypes compared to non-transgenic littermates. In vivo luciferase scanning of positive pTetOAROM mice did not detect any in vivo luciferase activity (data not shown), which confirmed that transgene expression under the control of the pTetO promoter is highly dependent on the presence of transactivator and inducing agent, doxycycline. This observation concurred with those of in vitro experiments—cells not expressing rtTA transactivator or not dox treated, could not express the pTetOdriven luciferase and human aromatase transgenes. The double transgenic mice (not treated with dox) resulted from the breeding of pTetOAROM mice with mammary-specific transactivator-expressing mice (MMTVrtTA), were also phenotypically normal compared to WT littermates. The expression of luciferase and human aromatase transgenes was tissue-specific and tightly regulated by dox. However, in vivo results from MMTVrtTApTetOAROM mice showed variations in transgene expression. Transgene expression was not detectable in every double MMTVrtTA-pTetOAROM mouse treated with dox. Moreover, double-transgenic mice within the same litter may display variable patterns of transgene expression. This may be due to the difference in the amount of dox-treated water consumed, and that different tissues may have varying degrees of sensitivity to dox. Dox injections may not be feasible, as the half-life of doxycycline in mice was found to be less than 3 h (Bo¨cker et al. 1981), hence repetitive injections may be required. For consistency, future studies with the tet-inducible systems should consider the use of subcutaneous implanted dox-releasing pellets, such that each animal will receive an equal amount of dox. However, the proportional amount of dox received using the pellet will be difficult to adjust to body weight. A previous study has found that doxycycline administration via feeds was the most effective compared to other non-invasive methods, especially with the drinking route which was associated with dehydration and weight loss (Cawthorne et al. 2007). The present study observed transgene expression predominantly in the salivary gland of dox-treated female and male double transgenic mice, and seminal

425

vesicle in males. This was particularly evident from in vivo luciferase scans of mice and tissue luciferase assays. Transgene expression was neither detectable by in vivo bioluminescence imaging or by RT-PCR in the mammary fat pad of dox-treated double transgenic female mice, but it was occasionally detected in male mammary fat pads. The mammary glandular development in male mice was arrested at birth hence only the rudimentary mammary bud remains within the pad. Therefore that may explain why the aromatase or luciferase expression in the male mammary gland was inconsistent and low, despite the fact that the rtTA transcript was clearly detectable in male ?/? mammary gland with or without dox treatment. Human aromatase and luciferase transcripts were readily detectable in the male salivary glands using real-time PCR. The lack of hArom and Luc transgenes expression in the (?/?) female mice was due to the absence of transactivator rtTA expression directed by the MMTV promoter. The mouse mammary tumor virus (MMTV) promoter was previously reported to be expressed in the mammary epithelium, as well as in many other tissues such as the epithelium of salivary glands, seminal vesicle and thymus (Ross et al. 1990). Hruska and colleagues (Hruska et al. 2002) showed that MMTV-tTA mediated overexpression of ERa transgene was detected in mammary gland, salivary gland, testis, seminal vesicle, and epididymis. Gunther and colleagues (Gunther et al. 2002) detected much stronger MMTVrtTA- and dox-induced pTetOlacZ expression in mammary glands of female double transgenic mice, and only mildly in the salivary gland, male seminal vesicles, and the thymus. Our transgenic male data showed that dox-treatment had no effect on rtTA expression, and hence rtTA expression in our current colony may interfer by other mechanisms such as epigenetic modification in the female animals. Nonetheless, these observations further emphasized that the presence of both rtTA and dox are crucial for the tight regulation of pTetOAROM transgene expression. Interestingly, doxycycline concentrations in liver and kidneys were previously observed to be greater in female compared to male (Bo¨cker et al. 1984). This may have affected the availability of doxycycline to other tissues (mammary and salivary glands) and hence the lower occurrence of transgene expression in female double transgenic mice.

123

426

Transgenic Res (2012) 21:415–428

Clearly, the redundancy and variability of tissuespecific promoter expression can be a limiting factor for generating conditional transgenic models. One possible resolution around this problem is to deliver transactivator expression using other methods such as virus-mediated expression, where only certain populations of cells will be infected. This method will be particularly useful for brain-specific induction of transgene expression when combining with the stereotaxic delivering method (Konopka et al. 2009), as well as for liver-specific induction (Connelly 1999). There had been a growing number of transgenic mouse models being generated to date, in order to study the role of aromatase and estrogen in different organs (Table 2). Our current model, pTetO-AROM, is an initial step towards an improved conditional aromatase expressing mouse model, where aromatase expression can be fine-tuned in terms of levels and timing, as well as in a tissue-specific manner (by cross-breeding pTetO-AROM with tissue-specific transactivator expressing mice as in the current study). The resultant transgenic mouse model will

also allow concurrent expression of a report protein (luciferase) to aid detection of transgene expression. The ultimate goal will be to cross-breed the tissuespecific pTetO-AROM mice with the global aromatase knockout mice. Therefore, the current model was not generated with the intention to overexpress aromatase/exaggerate estrogen production but to replace estrogen production in an estrogen-deficient mouse model. Therefore it is a unique aromatase transgenic model in itself. In summary, the present study demonstrates that the pTetOAROM construct can direct tissue-specific expression of human aromatase and luciferase as a marker gene, which was inducible by doxycycline (dox). Expression of pTetOAROM and its tissuespecificity was highly dependent on the presence of transactivator as well as the availability of doxycycline. Further studies can utilize this current pTetOAROM transgenic model to create tissue-specific aromatase-expressing mice by cross-breeding with transgenic mice expressing transactivator in any tissue, which may lead to the discovery of unexpected

Table 2 Models of aromatase transgenic models Transgenic model

Examples

Advantage

Tissue-specific aromatase overexpression

Mammary-specific MMTV-int/5 promoter-driven mouse aromatase cDNA expression (Tekmal et al. 1996)

To study the effect of increased aromatase expression at a particular tissue

Bone-specific rat type I a I procollagen promoterdriven human aromatase cDNA (Sjo¨gren et al. 2009) Aromatase tissuespecific promoterdriven reporter expression

Brain-specific aromatase promoter-driven bgalactosidase expression (Harada and Honda 2005)

To study the tissue-specificity of the multiple promoters of the aromatase gene

Adipose and ovary-specific aromatase promoterdrive human growth hormone expression (Hinshelwood et al. 2000, Hinshelwood and Mendelson 2001)

Global aromatase overexpression

Placenta-specific aromatase promoter-driven human growth hormone expression (Kamat et al. 1999) Promoter CMV-driven human aromatase cDNA expression (Li et al. 2001)

Tissue-specific expression of aromatase

Human aromatase cDNA driven by human aromatase brain-specific promoter (Harada et al. 2009)

To study the effect of brain aromatase expression

Conditional expression of aromatase

pTetO-AROM bi-transgenic (current project and future direction)

To replace aromatase expression at any particular tissue(s) whenever transgene is induced by an exogenous agent, concurrently with a reporter protein

123

To study the effect of increased aromatase throughout the body

Transgenic Res (2012) 21:415–428

roles of estrogens produced locally in various tissues of interest. Acknowledgments This work was supported by Australian NH&MRC Project Grant, #494813 to WC Boon, #395525 R Douglas Wright Fellowship to JT Price and NHMRC Equipment Grant #467202 to JT Price and by # 338510 and Program Grant # 441100, as well as a Program Grant from the Victorian Breast Cancer Research Consortium Inc to ER Simpson.

References Ackerman GE, Smith ME et al (1981) Aromatization of androstenedione by human adipose tissue stromal cells in monolayer culture. J Clin Endocrinol Metab 53:412–417 Albanese C, Reutens AT et al (2000) Sustained mammary gland-directed, ponasterone A-inducible expression in transgenic mice. Faseb J 14(7):877–884 Bo¨cker R, Estler CJ et al (1981) Comparison of distribution of doxycycline in mice after oral and intravenous application measured by a high-performance liquid chromatographic method. Arzneimittelforschung 31(12):2116–2117 Bo¨cker R, Warnke L et al (1984) Blood and organ concentrations of tetracycline and doxycycline in female mice. Comparison to males. Arzneimittelforschung 34(4):446–448 Boon WC, Diepstraten J et al (2005) Hippocampal NMDA receptor subunit expression and watermaze learning in estrogen deficient female mice. Mol Brain Res 140(1–2): 127–132 Bulun SE, Sebastian S et al (2003) The human CYP19 (aromatase P450) gene: update on physiologic roles and genomic organization of promoters. J Steroid Biochem Mol Biol 86(3–5):219–224 Carani C, Qin K et al (1997) Effect of testosterone and estradiol in a man with aromatase deficiency. N Engl J Med 337(2):91–95 Cawthorne C, Swindell R et al (2007) Comparison of doxycycline delivery methods for Tet-inducible gene expression in a subcutaneous xenograft model. J Biomol Tech 18(2):120–123 Chen D, Reierstad S et al (2009) Regulation of breast cancerassociated aromatase promoters. Cancer Lett 273(1): 15–27 Chow JDY, Simpson ER et al (2009) Alternative 50 -untranslated first exons of the mouse Cyp19A1 (aromatase) gene. J Steroid Biochem Mol Biol 115(3–5):115–125 Connelly S (1999) Adenoviral vectors for liver-directed gene therapy. Curr Opin Mol Ther 1(5):565–572 Demura M, Reierstad S et al (2008) Novel promoter I.8 and promoter usage in the CYP19 (Aromatase) gene. Reprod Sci 15(10):1044–1053 Gossen M, Bujard H (1992) Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci USA 89(12):5547–5551 Gunther EJ, Belka GK et al (2002) A novel doxycyclineinducible system for the transgenic analysis of mammary gland biology. FASEB J 16(3):283–292

427 Harada N, Honda S (2005) Analysis of spatiotemporal regulation of aromatase in the brain using transgenic mice. J Steroid Biochem Mol Biol 95(1–5):49–55 Harada N, Utsumi T et al (1993) Tissue-specific expression of the human aromatase cytochrome P-450 gene by alternative use of multiple exons 1 and promoters, and switching of tissue-specific exons 1 in carcinogenesis. Proc Natl Acad Sci USA 90(23):11312–11316 Harada N, Wakatsuki T et al (2009) Functional analysis of neurosteroidal oestrogen using gene-disrupted and transgenic mice. J Neuroendocrinol 21(4):365–369 Hinshelwood MM, Mendelson CR (2001) Tissue-specific expression of the human CYP19 (aromatase) gene in ovary and adipose tissue of transgenic mice. J Steroid Biochem Mol Biol 79(1–5):193–201 Hinshelwood MM, Smith ME et al (2000) A 278 bp region just upstream of the human CYP19 (aromatase) gene mediates ovary-specific expression in transgenic mice. Endocrinology 141(6):2050–2053 Hruska KS, Tilli MT et al (2002) Conditional over-expression of estrogen receptor alpha in a transgenic mouse model. Transgenic Res 11(4):361–372 Kamat A, Graves KH et al (1999) A 500-bp region, approximately 40 kb upstream of the human CYP19 (aromatase) gene, mediates placenta-specific expression in transgenic mice. Proc Natl Acad Sci USA 96(8):4575–4580 Konopka W, Duniec K et al (2009) Tet system in the brain: transgenic rats and lentiviral vectors approach. Genesis 47(4):274–280 Labbe GG, Fromenty BB et al (1991) Effects of various tetracycline derivatives on in vitro and in vivo beta-oxidation of fatty acids, egress of triglycerides from the liver, accumulation of hepatic triglycerides, and mortality in mice. Biochem Pharmacol 41(4):638–641 Lephart ED, Simpson ER (1991) Assay of aromatase activity. Methods Enzymol 206:477–483 Li X, Nokkala E et al (2001) Altered structure and function of reproductive organs in transgenic male mice overexpressing human aromatase. Endocrinology 142(6):2435–2442 Maffei L, Murata Y et al (2004) Dysmetabolic syndrome in a man with a novel mutation of the aromatase gene: effects of testosterone, alendronate, and estradiol treatment. J Clin Endocrinol Metab 89(1):61–70 Maffei L, Rochira V et al (2007) A novel compound heterozygous mutation of the aromatase gene in an adult man: reinforced evidence on the relationship between congenital oestrogen deficiency, adiposity and the metabolic syndrome. Clin Endocrinol (Oxf) 67(2):218–224 Mahendroo MS, Mendelson CR et al (1993) Tissue-specific and hormonally controlled alternative promoters regulate aromatase cytochrome P450 gene expression in human adipose tissue. J Biol Chem 268(26):19463–19470 McNatty KP, Baird DT et al (1976) Concentration of oestrogens and androgens in human ovarian venous plasma and follicular fluid throughout the menstrual cycle. J Endocrinol 71(1):77–85 Morishima A, Grumbach MM et al (1995) Aromatase deficiency in male and female siblings caused by a novel mutation and the physiological role of estrogens. J Clin Endocrinol Metab 80(12):3689–3698

123

428 Robertson KM, O’Donnell L et al (1999) Impairment of spermatogenesis in mice lacking a functional aromatase (cyp 19) gene. PNAS 96(14):7986–7991 Ross SR, Hsu CL et al (1990) Negative regulation in correct tissue-specific expression of mouse mammary tumor virus in transgenic mice. Mol Cell Biol 10(11):5822–5829 Sasano H, Nagura H et al (1994) Immunolocalization of aromatase and other steroidogenic enzymes in human breast disorders. Hum Pathol 25(5):530–535 Simpson ER (2003) Sources of estrogen and their importance. J Steroid Biochem Mol Biol 86(3–5):225–230 Simpson ER, Evans CT et al (1987) Sequencing of cDNA inserts encoding aromatase cytochrome P-450 (P-450AROM). Mol Cell Endocrinol 52(3):267–272 Sjo¨gren K, Lagerquist M et al (2009) Elevated aromatase expression in osteoblasts leads to increased bone mass without systemic adverse effects. J Bone Miner Res 24(7):1263–1270

123

Transgenic Res (2012) 21:415–428 Tekmal RR, Ramachandra N et al (1996) Overexpression of int-5/aromatase in mammary glands of transgenic mice results in the induction of hyperplasia and nuclear abnormalities. Cancer Res 56(14):3180–3185 Yamamoto N, Christenson LK et al (2002) Growth differentiation factor-9 inhibits 30 50 -adenosine monophosphatestimulated steroidogenesis in human granulosa and theca cells. J Clin Endocrinol Metab 87(6):2849–2856 Yarranton GT (1992) Inducible vectors for expression in mammalian cells. Curr Opin Biotechnol 3(5):506–511 Zhang N, Weber A et al (2003) An inducible nitric oxide synthase-luciferase reporter system for in vivo testing of anti-inflammatory compounds in transgenic mice. J Immunol 170(12):6307–6319 Zhou J, Suzuki T et al (2005) Interactions between prostaglandin E2, liver receptor homologue-1, and aromatase in breast cancer. Cancer Res 65(2):657–663