Develop. Growth Differ.

(2003)

45

, 377–387

Physiological regulation of P-glycoprotein, MRP1, MRP2 and cytochrome P450 3A2 during rat ontogeny

Anna Rosati, Silvia Maniori, Giuliana Decorti,* Luigi Candussio, Tullio Giraldi and Fiora Bartoli

Department of Biomedical Sciences, University of Trieste, Via L. Giorgieri no. 7, 34100 Trieste, Italy

P-glycoprotein and the multidrug resistance-related proteins MRP1 and MRP2 belong to the ATP bindingcassette family of proteins and transport a wide range of substrates. These proteins are also involved inmetabolic and excretory processes of xenobiotics. The rat genes

mdr1a

and

mdr1b

code for P-glycoproteins,while

mrp1

and

mrp2

genes code for MRP1 and MRP2 proteins, respectively. In this study, the physiologicalmodulation of the level of transcript for these genes during rat ontogeny in the liver, kidney, lung, brain and heartwas analyzed by reverse transcription–polymerase chain reaction. An increasing level of transcript duringontogeny was demonstrated for

mdr1a

and

mdr1b

in all tissues considered, as well as for

mrp2

, which wasdetected only in the liver and kidney. In contrast,

mrp1

transcript, present in all tissues, did not show anymodulation. The maximum level of expression was reached in adult animals and a significant decrease wasdemonstrated in aging rats. Western blot analysis with C219 and M

2

III-6 monoclonal antibodies confirmed thisdifferent pattern of expression during ontogeny in the liver. The physiological regulation of cytochromeP450 3A2 was also considered: in the rat liver, an increase in the level of transcript during ontogeny, with amaximum in 60-day-old rats and a decrease in 8-month-old rats, was evident.

Key words:

multidrug resistance-related protein, P-glycoprotein, RT–PCR, western blotting.

Introduction

P-glycoprotein and multidrug resistance related-proteins (MRP) are members of the ATP bindingcassette superfamily of transporters: these proteinsare located on the cellular membrane and activelypump a wide range of substrates out of the cell.P-glycoprotein is a 170 kDa protein coded in humansby the

MDR1

gene, located on chromosome 7. Inrodents, two isoforms have been described that arecoded by

mdr1a

and

mdr1b

genes. An additionalisoform, MDR3 in humans and mdr2 in rodents,expressed predominantly in the liver, does not confermultidrug resistance and appears to be involved in thetransport of phospholipids into the bile (Smit

et al

.1994; Crawford

et al

. 1997; Borst

et al

. 2000b). MRPare a group of nine members: the most studied, MRP1and MRP2, are proteins of approximately 190 kDa,and are coded for by

MRP1

and

MRP2

genes inhumans and by

mrp1

and

mrp2

genes in rodents.These proteins play a crucial role in multidrug

resistance in tumor cells, where their increased

expression is responsible for the failure of therapy.They are also widely distributed in normal tissues.P-glycoprotein is mainly expressed in the intestine,adrenal gland, kidney and liver. MRP1 is ubiquitouslydistributed throughout the body, with high levels inthe kidney, liver, gut and placenta, as well as in thelung, heart and blood–brain barrier (Borst

et al

. 1999,2000a). MRP2 is described as a strictly localizedprotein and is predominantly expressed at the biliarypole of the hepatocyte, but it has also been found inthe kidney and gut (Borst

et al

. 1999, 2000a).The presence of these proteins in normal tissues

suggests an important physiological role. In particular,their location in the liver, kidney, intestine, blood–brainbarrier and placenta suggests a secretory functionand a role in detoxification processes and in pro-tecting normal tissues from endogenous and exo-genous toxic substances (Ayrton & Morgan 2001).These proteins can be induced by their substrates,and induction by the substrates occurs both in tumorcells, where it is responsible for resistance, and innormal tissues, where it causes alterations in drugpharmacokinetics (Yu 1999).

Recent studies have revealed a great overlapbetween substrates of P-glycoprotein and CYP 3A4(Wacher

et al

. 1995; Schuetz

et al

. 1996; Yu 1999), thecytochrome isoform involved in the metabolism of

*Author to whom all correspondence should be addressed.Email: decorti@units.itReceived 24 January 2003; revised 7 May 2003; accepted

8 May 2003.

378 A. Rosati

et al.

most drugs. A different drug accumulation in tissues,indeed, is the result of a synergistic effect betweenthe transport proteins and cytochromes (Benet &Cummins 2001).

Physiological modulation of the

mdr1a

and

mdr1b

genes in mice during ontogeny has been described(Dutt

et al

. 1992; Schiengold

et al

. 2001), and thelevels of P-glycoprotein increase depending on thepostnatal day. The levels of cytochromes are regu-lated during ontogeny in humans (Hines & McCarver2002).

In this study, we further analyzed the physiologicalmodulation of the levels of

mdr1a

and

mdr1b

tran-scripts by reverse transcription (RT)–polymerasechain reaction (PCR) analysis in several rat tissuesfrom fetal, neonatal, and young and old adult rats,paying particular attention to the first week of life andthe aging period. In addition, we extended our studyto other transporters (MRP1 and MRP2), and alsoconsidered the expression of cytochrome P450 3A2.This isoform is the rat homologue of human CYP4503A4 (Durr

et al

. 2000), the most important isoform ofdrug metabolizing enzymes in humans. In addition,western blotting with C219 anti-P-glycoprotein andM

2

III-6 anti-MRP2 monoclonal antibodies was alsocarried out in the liver.

Materials and Methods

Chemicals

Mouse anti-IgG alkaline phosphatase-conjugatedantibody, RedTaq DNA polymerase and a 100 bpDNA ladder were purchased from Sigma-Aldrich(Milan, Italy). SV Total RNA Isolation System wasfrom Promega (Milan, Italy); SuperScript II RNaseH

–

reverse transcriptase and synthetic oligonucleotideprimers were from Invitrogen Life Technologies (SanGiuliano Milanese, Italy); and C219 and M

2

III-6 mono-clonal antibodies were from Alexis Vinci Biochem

(Vinci, Italy). All other chemicals were of analyticalgrade.

Animals

Wistar rats were obtained from a local conventionalbreeding colony. The animals, housed in standardcages at 20

�

C on a 12 h light : dark cycle, had freeaccess to both food and water. Animal experimentswere carried out in accordance with Italian laws (Min-istry of Health registration no. 62/2000-B, 6 October2000).

At predetermined time intervals, the animals werekilled by decapitation after ether anesthesia. Tissuesfrom six animals at postnatal day 0, 2, 4, 6, 12, 20, 30and 60, and from three animals 5- and 8-month-oldwere obtained. Three pregnant rats were subjected tocesarean section after anesthesia to obtain 15-day-old fetuses, which were separated from placentas,and their tissues were excised under the microscope.The tissues obtained from each litter were pooled.

Reverse transcription–polymerase chain reaction

Total RNA was extracted with the SV Total RNA Iso-lation System from tissues collected on ice and storedat –80

�

C. Total RNA (5 µg) was reverse transcribedusing Superscript II RNase H

–

and the cDNA was

Table 1.

Primers used in the polymerase chain reaction

Gene Strand Sequence Accession number Length (bp)

mdr1a

Sense 5

�

-ACTCGGGAGCAGAAGTTTGA-3

�

Antisense 5

�

-GGAGCCACTGGACATTGAGT-3

�

AF257746 600

mdr1b

Sense 5

�

-GACGGACAGGACATCAGGACC-3

�

Antisense 5

�

-GCAAACACTGGTTGTATGCAC-3

�

M62425 823

mrp1

Sense 5

�

-CCATTCAGGCCGTGTAGAGT-3

�

Antisense 5

�

-TCATGGTTCAGCTTGTCAGG-3

�

AJ277881 404

mrp2

Sense 5

�

-CCACAGGCCGGATTGTG-3

�

Antisense 5

�

-AAGATTCTGAAGAGGCAG-3

�

D86086 826

CYP 3A2

Sense 5

�

-AGTGGGGATTATGGGGAAAG-3

�

Antisense 5

�

-AATCCTTTGGGAACATGCAG-3

�

U09742 406

�

-actin

Sense 5

�

-CACCAAAGCTGAGAGGGAAATC-3

�

Antisense 5

�

-ATTTGCGGTGCACGATGGAGGGG-3

�

M10277 500

Fig. 1.

Level of expression of

mdr1a

in rat liver (a), kidney (b),heart (c), brain (d) and lung (e). The histograms represent themean

±

SE of the ratio obtained from densitometric bandquantification of specific primers and that of

�

-actin, expressedas percentage of the increase calculated versus the data of dayzero. A representative electrophoretic run of polymerase chainreaction products obtained with specific and

�

-actin primers isalso shown. Statistical analysis was done by

ANOVA

withDunnett’s post test. Significantly different from data obtained atday zero at *

P

< 0.05; **

P

< 0.01; significantly different from dataobtained at day 60 at

�

P

< 0.05;

��

P

< 0.01.

Ontogeny of MDR proteins in the rat 379

380 A. Rosati

et al.

used in PCR in a final volume of 10 µL. PCR wascarried out using serial dilution of cDNA according toBates

et al

. (1999).Synthetic oligonucleotide primers were used to

study the expression of

mdr1a

,

mdr1b

,

mrp1

,

mrp2

and

CYP 3A2

mRNA transcripts. The primer sequ-ences used for PCR analysis are summarized inTable 1.

The amplification by

Taq

polymerase was carriedout using a DNA thermal cycler Gene Amp 9700 (PEApplied Biosystems, Foster City, CA, USA). After a firstdenaturing step at 94

�

C for 5 min, the reaction wasconducted for 20–35 cycles; each cycle consisting of1 min at 94

�

C, 1 min at 60

�

C and 1 min at 72

�

C. For allprimers, the reaction ended with a final step at 72

�

Cfor 10 min. PCR products were subjected to electro-phoresis on a 1.5% agarose gel and visualized byethidium bromide staining. The DNA marker was a100 bp ladder. Primers amplifying a fragment of

�

-actin were also used.Polymerase chain reaction products, obtained by

the amplification of serially diluted aliquots of cDNA,were densitometrically quantified by GelCompar II(Applied Maths, Kortrijk, Belgium). To normalize data,the ratio between the densitometric quantificationobtained with specific primers and with

�

-actinprimers was calculated (Bates

et al

. 1999).

Western blot analysis

Livers from animals at day 0, 15, 60 and 8 monthswere homogenized with Ultra Turrax (IKA-Werk,Staufen, Germany) in 20 volumes of lysis buffer(100 m

M

NaCl, 1 m

M

ethylenediaminetetraacetic acid(EDTA) and 10 m

M

Tris-HCl (pH 7.5)), protease inhib-itors were added (1 m

M

phenylmethanesulfonyl fluo-ride, 2 µg/mL aprotinin, 2 µg/mL leupeptin and1 µg/mL pepstatin A), then the samples werecentrifuged at 4

�

C for 10 min at 1000

g

. After proteinquantification using the Bio-Rad Kit (Bio-Rad Labor-atories, Segrate, Italy), the supernatant was aliquotedand stored at –20

�

C until use. Samples were diluted at50 µg protein in loading buffer (Laemmli 1970), boiledfor 5 min, centrifuged for 5 min at 15 700 g, thenloaded on 8% sodium dodecylsulfate (SDS)–poly-acrylamide gel electrophoresis (PAGE) gels. Theresolved proteins were transferred to a nitrocellulosemembrane by the semi-dry protein blotting system(LKB Nova Blot; Pharmacia Biotech, ColognoMonzese, Italy) at 2 mA/cm2 for 2 h.

The membrane was stained with ponceau red toverify the homogeneity of samples, then blocked for atleast 2 h in Tris-buffered saline (150 mM NaCl, 50 mM

Tris HCl (pH 7.5)) containing 0.05% v/v Tween-20

(TTBS) with 3% skim milk added, and incubatedovernight with C219 (1:200 in TTBS with 3% milk) orM2III-6 (1:50 in TTBS with 3% milk) antibodies. Blotswere washed three times with TTBS, and incubatedwith an antimouse secondary antibody conjugatedwith alkaline phosphatase (1:30 000 in TTBS with 3%milk) for 1 h at room temperature. The membrane wasextensively washed and finally stained by addition ofbromocloroindolyl phospatase and nitroblue tetra-zolium, two substrates for alkaline phosphatase.

Statistical analysis

Data obtained by PCR band quantification were com-pared by one-way ANOVA followed by Dunnett’s posttest if ANOVA reached statistical significance.

Results

Reverse transcription–polymerase chain reaction

Mdr1a This transcript was detected, by RT–PCR, inall tissues considered. Physiological regulation of thelevel of expression of mdr1a transcript was evident,with a significant increase on day six in the kidney(P < 0.05), day 20 in the lung (P < 0.01) and liver(P < 0.01), and day 30 in the heart (P < 0.05) andbrain (P < 0.05).

A significant decrease in the level of this transcriptwas evident in the brain (P < 0.05) of 5-month-old ratsand in the liver (P < 0.01) and heart (P < 0.05) of8-month-old rats, compared with that of 60-day-oldanimals. In other tissues, no difference was detected(Fig. 1).

Mdr1b This transcript was detected in all tissuesconsidered (Fig. 2). A significant increase was evidenton postnatal day six in the liver (P < 0.05), kidney(P < 0.01) and heart (P < 0.05), day 12 in the brain(P < 0.01), and day 20 in the lung (P < 0.01). Similarlyto mdr1a, the maximum level of mdr1b transcript wasreached by postnatal day 60.

Fig. 2. Level of expression of mdr1b in rat liver (a), kidney (b),heart (c), brain (d) and lung (e). The histograms represent themean ± SE of the ratio obtained from densitometric bandquantification of specific primers and that of �-actin, expressedas percentage of increase calculated versus the data of dayzero. A representative electrophoretic run of polymerase chainreaction products obtained with specific and �-actin primers isalso shown. Statistical analysis was done by ANOVA withDunnett’s post test. Significantly different from data obtained atday zero at *P < 0.05; **P < 0.01; significantly different from dataobtained at day 60 at ��P < 0.01.

Ontogeny of MDR proteins in the rat 381

382 A. Rosati et al.

Ontogeny of MDR proteins in the rat 383

In contrast, during aging, a significant decreasein mdr1b transcript (P < 0.01) was observed in alltissues obtained from 5- and 8-month-old versus60-day-old rats.

Mrp1 The mrp1 transcript was also monitored(Fig. 3). With our PCR primers, it was not possible toshow any variation during ontogenesis in all tissuesconsidered. Only in the liver of 8-month-old rats andthe kidney of 5- and 8-month-old-rats, the transcriptwas significantly lower (P < 0.01) in comparison withthe adult of 60 days.

Mrp2 Of the tissues examined, the presence of mrp2transcript was demonstrated with our primers in theliver and kidney, and a physiological regulationduring ontogenesis was evident (Fig. 4). In com-parison with day zero data, a significant increase ofthis transcript was observed in the liver, beginning at

postnatal day six (P < 0.05) and increasing untilday 30 (P < 0.01). In this tissue, the level of mrp2transcript was significantly reduced (P < 0.01) in8-month-old rats in comparison with day 60. Similarly,in the kidney, we demonstrated a significant increasefrom postnatal day 12 (P < 0.05), a peak at day 60(P < 0.01) and a decrease in 5- and 8-month-old rats(P < 0.01).

CYP 3A2 The transcript for CYP 3A2 was detected inthe liver, where it showed a physiological increaseduring development (P < 0.05 at postnatal day two)with the maximum level at day 60 (Fig. 5). In 5- and8-month-old rats, the transcript had significantlydecreased (P < 0.01) in comparison with the levelobserved in adults of 60 days. In the kidney, ourprimers recognized this transcript but the results didnot indicate any regulation during rat ontogeny.

Western blot

C219 anti-P-glycoprotein monoclonal antibody recog-nized a band of 170 kDa in livers obtained from15- and 60-day-old rats; the intensity was higher in60-day-old rats. No signal was evident in tissues from0 day and 8-month-old animals (Fig. 6). M2III-6 anti-MRP2 monoclonal antibody detected a 190 kDa bandin the livers obtained from 15- and 60-day-old and8-month-old rats, but not in those from newbornanimals. The intensity was higher in young animals,and decreased in older animals (Fig. 7).

Fig. 3. Level of expression of mrp1 in rat liver (a), kidney (b),heart (c), brain (d) and lung (e). The histograms represent themean ± SE of the ratio obtained from densitometric bandquantification of specific primers and that of �-actin, expressedas percentage of increase calculated versus the data of dayzero. A representative electrophoretic run of polymerase chainreaction products obtained with specific and �-actin primers isalso shown. Statistical analysis was done by ANOVA withDunnett’s post test. Significantly different from data obtained atday zero at **P < 0.01; significantly different from data obtainedat day 60 at ��P < 0.01.

Fig. 4. Level of expression of mrp2 in rat liver (a) and kidney (b). The histograms represent the mean ± SEM of the ratio obtained fromdensitometric band quantification of specific primers and that of �-actin, expressed as percentage of increase calculated versus thedata of day zero. A representative electrophoretic run of polymerase chain reaction products obtained with specific and �-actin primersis also shown. Statistical analysis was done by ANOVA with Dunnett’s post test. Significantly different from data obtained at day zero at*P < 0.05; **P < 0.01; significantly different from data obtained at day 60 at ��P < 0.01.

384 A. Rosati et al.

Discussion

During the prenatal and postnatal periods, manyorgans are incompletely developed, and manyprotective enzyme systems undergo substantial andrapid changes in activity. Therefore, during develop-ment, important modifications in the pharmacokineticsof many drugs occur, and the consequence of this isdifferences in therapeutic efficacy and toxicity. Thesechanges have usually been ascribed to the ontogenicdevelopment of drug metabolizing enzymes; however,recent studies have shown that other proteins, inparticular P-glycoprotein and MRP, can also beinvolved in drug pharmacokinetics (Yu 1999). Theseproteins are responsible for the phenomenon ofmultidrug resistance in tumor cells, but are alsoexpressed in many normal tissues. During recentyears, developmental changes in their expressionhave been described; however, results obtained withdifferent techniques by various authors have oftenbeen conflicting (Dutt et al. 1992; Matsuoka et al.1999; Mahmood et al. 2001; Mottino et al. 2001;Schiengold et al. 2001; Zinchuk et al. 2002).

In this study, the physiological changes of mdr1a,mdr1b, mrp1, mrp2 and CYP 3A2 transcript levelswere evaluated by RT–PCR, and PCR products,obtained from serially diluted cDNA, were densito-

metrically quantified as described by Bates et al.(1999). Tissues with a metabolic and an excretoryfunction, such as the liver and kidney, but also thebrain, heart and lung, were collected from rats atdifferent developmental stages. The intestine was alsoanalyzed, but only a faint signal was obtained. Thisresult can probably be ascribed to the type of prepar-

Fig. 5. Level of expression of CYP 3A2 in rat liver. The histo-gram represents the mean ± SEM of the ratio obtained fromdensitometric band quantification of specific primers and that of�-actin, expressed as percentage of increase calculated versusthe data of day zero. A representative electrophoretic run ofpolymerase chain reaction products obtained with specific and�-actin primers is also shown. Statistical analysis was done byANOVA test with Dunnett’s post test. Significantly different fromdata obtained at day zero at *P < 0.05; **P < 0.01; significantlydifferent from data obtained at day 60 at ��P < 0.01.

Fig. 6. Western blot analysis of P-glycoprotein expression inrat liver by C219 monoclonal antibody. A representativeimmunoblot is shown in (A). Proteins (50 µg) extracted from theliver of animals of different ages were loaded on each lane. C219antibody (1:200) was incubated as described in the text. In (B),the reversible staining of the same membrane with ponceau redis shown to verify that a comparable quantity of protein waseffectively loaded on the gel.

Fig. 7. Western blot analysis of MRP2 expression in rat liver byM2III-6 monoclonal antibody. A representative immunoblot isshown in (A). Proteins (50 µg) extracted from the liver of animalsof different ages were loaded on each lane. M2III-6 antibody(1:50) was incubated as described in the text. In (B), thereversible staining of the same membrane with ponceau red isreported to verify that a comparable quantity of protein waseffectively loaded on the gel.

Ontogeny of MDR proteins in the rat 385

ation: we used the first part of the whole intestine,whereas transport proteins have a strictly apicallocalization on the villi and hence a preparationenriched in this fraction would probably give betterresults (Mahmood et al. 2001; Mottino et al. 2001).

The transcripts for mdr1a and mdr1b were detectedin all organs examined, in agreement with the otherstudies on P-glycoprotein tissue localization in rodents(Croop et al. 1989), and a physiological regulationduring rat ontogenesis was observed. High levelswere reached in all tissues examined within ratweaning and sexual maturity; however, during aging,a significant decrease was observed in almost allorgans considered. Similar results were obtained byDutt et al. (1992). Using RNase protection assay andimmunohistochemistry, these authors demonstrated,in mice from birth to 6 weeks, an increase of mdr1aand mdr1b in the kidney and liver, the highest levelsbeing reached between the third and sixth week ofage. In contrast, Mahmood et al. (2001) were not ableto show any change in P-glycoprotein expression inthe liver and kidney of mice from day zero to adult-hood, whereas a significant increase in P-glycoproteinexpression with maturation was demonstrated in theintestine.

Multidrug resistance proteins are also expressed inmany normal tissues, and a physiological modulationduring ontogenesis can also be hypothesized forthese proteins. Transcript for mrp2 was detected onlyin the liver and kidney, and its developmental behaviorwas similar to that described above for the mdr1a andmdr1b transcripts. In fact, the level of mrp2 signifi-cantly increased during development in animals of1 week (liver) or 2 weeks (kidney) of age. Theseresults agree with the recent observation of Zinchuket al. (2002), who demonstrated an increase of mrp2transcript in the rat liver during ontogeny. Moreover,we demonstrated that the same regulation was alsopresent in the kidney and in addition, also for thisprotein, a decrease was observed in the aging rat incomparison with the young animal.

In our experimental conditions, MRP1, detected inall tissues examined, showed no increase during ratmaturation. These results might suggest a differentrole of the MRP1 and MRP2 proteins: MRP2 seems tobe involved in protection against xenobiotics and theirdetoxification, being present predominantly in meta-bolic and excretory tissues such as the liver andkidney; and MRP1 might play a physiological role inthe transport of endogenous compounds crucial forcellular survival and for cellular communication, and istherefore expressed from the early phases of life.These proteins have a different cellular localization(MRP2 is located on the apical side while MRP1 is

basolateral), and have a different tissue distribution,MRP1 being more ubiquitous than MRP2 (Borst et al.1999, 2000a).

P-glycoprotein, MRP1 and MRP2 act in synergy withcytochromes to protect the organism from toxiccompounds (Benet & Cummins 2001); therefore, weexamined the level of transcription of cytochromeP450 3A2. This isoform is the homologue of humancytochrome P450 3A4 (Durr et al. 2000), which is themain enzyme involved in the metabolism of toxicsubstances that are also substrates of P-glycoprotein,MRP1 and MRP2 (Guengerich 1999). A physiologicalregulation of cytochromes during ontogeny in humansis well known (Hines & McCarver 2002), and our dataconfirm the regulation of cytochrome P450 3A2 in therat liver during ontogenesis. The transcript was alsodetectable in the kidney and intestine, but in these twotissues it was not possible to demonstrate any regu-lation. In humans, regulation of CYP450 3A4 hasbeen described in these tissues; however, in the rat,besides CYP450 3A2, there are other isoforms withthe same function that show different tissue distrib-ution and sex specificity (Johnson et al. 2000; Kawaiet al. 2000). Our study confirms the developmentalchanges of CYP 3A2 in the liver, but also suggests animportant role in the detoxifying processes duringontogenesis for P-glycoprotein and MRP2, hence thedisposition and liver excretion of many drugs andxenobiotics seems to be regulated by the dynamicbalance between these enzymes and transportproteins. The perinatal kidney is also incompletelydeveloped, and drug transporters undergo importantchanges in activity during this period (Klinger 1982).We demonstrated that mdr1, mdr2 and mrp2 geneexpression increases during maturation in the kidney,and is not complete until day 60. In addition, inthe aging animal, gene expression significantlydecreases, possibly contributing to the decline inrenal drug excretion observed in the older rats.

A physiological regulation in mdr1a and mdr1b wasobserved in our study in the brain. P-glycoprotein islocated on the luminal surface of endothelial cells ofthe blood brain capillaries (Thiebaut et al. 1989;Stewart et al. 1996; Matsuoka et al. 1999), and isbelieved to act as a part of the blood–brain barrier,playing a role in limiting the access of manycompounds into the cerebral parenchyma. Disruptionof P-glycoprotein function is indeed responsible forincreased toxicity (Mayer et al. 1996; Schinkel et al.1994; Schinkel et al. 1996) and altered pharmaco-logical activities of many drugs in the central nervoussystem (Schinkel et al. 1996). The blood–brain barrieris incomplete in the embryo and newborn and, as aconsequence, the neurological toxicity of many sub-

386 A. Rosati et al.

stances is increased during this age period. Indeed,in our study, the expression of mdr1a and mdr1bgenes in the central nervous system rapidly increasedafter birth, in accord with Matsuoka et al. (1999), andreached a maximum at approximately day 30 formdr1a and day 12 for mdr1b. Based on these find-ings, and in combination with results from previousreports, an important role of these proteins in thedevelopment and maturation of the blood–brainbarrier can therefore be suggested. The level ofexpression of both mdr1a and mdr1b genesdecreased significantly at 5 and 8 months. Malfunc-tion of the blood–brain barrier has been observedduring the process of senescence (Shah & Moor-adian 1997), and, among the possible causes of theincreased permeability during aging, a reduction inP-glycoprotein expression could be suggested.

An increase of mdr1a and mdr1b transcripts wasobserved also in tissues not involved in metabolismand excretion, such as the heart and lung, suggestingthat these transporters play a physiological role in theprotection of these tissues against toxins and exogen-ous substances, when, for example, the rat becomesindependent from breast feeding. The factors respon-sible for these changes are unclear, but can includeendogenous and exogenous compounds. It is indeedknown that the expression of MDR1 and MRP2 isinduced by steroids and other hormones whoseactivity may change during aging (Courtois et al.1999; Demeule et al. 1999). In addition, breast milkcontains a number of growth factors and hormonesthat are known to stimulate P-glycoprotein expressionand function in tumor cell lines in vitro (Yang et al.1997; Guo et al. 1998).

In conclusion, this study has demonstrated a coor-dinate modification in the expression of transport pro-teins and of CYP 3A2 in the prenatal and postnatalperiods, as well as during aging. These alterationscould be responsible for the important and rapidchanges in the pharmacokinetics of many drugsobserved during these periods of life.

Acknowledgements

This research was supported by grants from the Minis-tero Università e Ricerca Scientifica e Tecnologica(60%), Programmi di Ricerca Scientifica di RilevanteInteresse Nazionale (Farmacotossicologia di antracic-line antitumorali: Meccanismi e basi molecolari pernuove strategie terapeutiche).

References

Ayrton, A. & Morgan, P. 2001. Role of transport proteins in drugabsorption, distribution and excretion. Xenobiotica 31,469–497.

Bates, S. E., Zhan, Z. R., Regis, J. & Gamelin, E. 1999.Measuring MDR-1 by quantitative RT-PCR. In CytotoxicDrug Resistance Mechanisms (Eds R. Brown & U. Böger-Browns), pp. 63–81. Humana Press, Totowa.

Benet, L. Z. & Cummins, C. L. 2001. The drug efflux-metabolismalliance: biochemical aspects. Adv. Drug Deliv. Rev. 50,S3–S11.

Borst, P., Evers, R., Kool, M. & Wijnholds, J. 1999. The multidrugresistance protein family. Biochim. Biophys. Acta Biomem-branes 1461, 347–357.

Borst, P., Evers, R., Kool, M. & Wijnholds, J. 2000a. A family ofdrug transporters: The multidrug resistance-associatedproteins. J. Natl Cancer Inst. 92, 1295–1302.

Borst, P., Zelcer, N. & van Helvoort, A. 2000b. ABC transportersin lipid transport. Biochim. Biophys. Acta Mol. Cell. Biol.Lipids 1486, 128–144.

Courtois, A., Payen, L., Guillouzo, A. & Fardel, O. 1999. Up-regulation of multidrug resistance-associated protein 2(MRP2) expression in rat hepatocytes by dexamethasone.FEBS Lett. 459, 381–385.

Crawford, A. R., Smith, A. J., Hatch, V. C., Elferink, R. P. J. O.,Borst, P. & Crawford, J. M. 1997. Hepatic secretion ofphospholipid vesicles in the mouse critically depends onmdr2 or MDR3 P-glycoprotein expression – Visualization byelectron microscopy. J. Clin. Invest. 100, 2562–2567.

Croop, J. M., Raymond, M., Haber, D. et al. 1989. The threemouse multidrug resistance (mdr) genes are expressed ina tissue-specific manner in normal mouse tissues. Mol. Cell.Biol. 9, 1346–1350.

Demeule, M., Jodoin, J., Beaulieu, E., Brossard, M. & Beliveau,R. 1999. Dexamethasone modulation of multidrug trans-porters in normal tissues. FEBS Lett. 442, 208–214.

Durr, D., Stieger, B., Kullak-Ublick, G. A. et al. 2000. St John’sWort induces intestinal P-glycoprotein/MDR1 and intestinaland hepatic CYP3A4. Clin. Pharmacol. Ther. 68, 598–604.

Dutt, A., Priebe, T. S., Teeter, L. D., Kuo, M. T. & Nelson, J. A.1992. Postnatal development of organic cation transportand mdr gene expression in mouse kidney. J. Pharmacol.Exp. Ther. 261, 1222–1230.

Guengerich, F. P. 1999. Cytochrome P-450 3A4: regulation androle in drug metabolism. Ann. Rev. Pharmacol. Toxicol. 39,1–17.

Guo, Y. S., Jin, G. F., Houston, C. W., Thompson, J. C. &Townsend, C. M. 1998. Insulin-like growth factor-I promotesmultidrug resistance in MCLM colon cancer cells. J. Cell.Physiol. 175, 141–148.

Hines, R. N. & McCarver, D. G. 2002. The ontogeny of humandrug-metabolizing enzymes: Phase I oxidative enzymes.J. Pharmacol. Exp. Ther. 300, 355–360.

Johnson, T. N., Tanner, M. S. & Tucker, G. T. 2000. A comparisonof the ontogeny of enterocytic and hepatic cytochromesP450, 3A in the rat. Biochem. Pharmacol. 60, 1601–1610.

Kawai, M., Bandiera, S. M., Chang, T. K. & Bellward, G. D. 2000.Growth hormone regulation and developmental expressionof rat hepatic CYP3A18, CYP3A9, and CYP3A2. Biochem.Pharmacol. 59, 1277–1287.

Klinger, W. 1982. Biotransformation of drugs and other xeno-biotics during postnatal development. Pharmac. Ther. 16,377–429.

Laemmli, U. K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 227,680–685.

Mahmood, B., Daood, M. J., Hart, C., Hansen, T. W. & Watchko,J. F. 2001. Ontogeny of P-glycoprotein in mouse intestine,liver, and kidney. J. Invest. Med. 49, 250–257.

Ontogeny of MDR proteins in the rat 387

Matsuoka, Y., Okazaki, M., Kitamura, Y. & Taniguchi, T. 1999.Developmental expression of P-glycoprotein (Multidrugresistance gene product) in the rat brain. J. Neurobiol. 39,383–392.

Mayer, U., Wagenaar, E., Beijnen, J. H. et al. 1996. Substantialexcretion of digoxin via the intestinal mucosa and pre-vention of long-term digoxin accumulation in the brainby the mdr1a P-glycoprotein. Br. J. Pharmacol. 119,1038–1044.

Mottino, A. D., Hoffman, T., Jennes, L., Cao, J. & Vore, M. 2001.Expression of multidrug resistance-associated protein 2in small intestine from pregnant and postpartum rats.Am. J. Physiol. Gastrointest. Liver Physiol. 280, G1261–G1273.

Schiengold, M., Schwantes, L., Schwartsmann, G., Chies, J. A.& Nardi, N. B. 2001. Multidrug resistance gene expressionduring the murine ontogeny. Mech. Ageing Dev. 122,255–270.

Schinkel, A. H., Smit, J. J., van Tellingen, O. et al. 1994. Dis-ruption of the mouse mdr1a P-glycoprotein gene leads to adeficiency in the blood–brain barrier and to increasedsensitivity to drugs. Cell 77, 491–502.

Schinkel, A. H., Wagenaar, E., Mol, C. A. A. M. & vanDeemter,L. 1996. P-glycoprotein in the blood–brain barrier of miceinfluences the brain penetration and pharmacologicalactivity of many drugs. J. Clin. Invest. 97, 2517–2524.

Schuetz, E. G., Beck, W. T. & Schuetz, J. D. 1996. Modulatorsand substrates of P-glycoprotein and cytochrome P4503Acoordinately up-regulate these proteins in human coloncarcinoma cells. Mol. Pharmacol. 49, 311–318.

Shah, G. N. & Mooradian, A. D. 1997. Age-related changes inthe blood–brain barrier. Exp. Gerontol. 32, 501–519.

Smit, J. J., Schinkel, A. H., Mol, C. A. et al. 1994. Tissuedistribution of the human MDR3 P-glycoprotein. LaboratoryInvest. 71, 638–649.

Stewart, P. A., Beliveau, R. & Rogers, K. A. 1996. Cellularlocalization of P-glycoprotein in brain versus gonadalcapillaries. J. Histochem. Cytochem. 44, 679–685.

Thiebaut, F., Tsuruo, T., Hamada, H., Gottesman, M. M., Pastan,I. & Willingham, M. C. 1989. Immunohistochemical localiz-ation in normal tissues of different epitopes in the multidrugtransport protein P170: evidence for localization in braincapillaries and crossreactivity of one antibody with a muscleprotein. J. Histochem. Cytochem. 37, 159–164.

Wacher, V. J., Wu, C. Y. & Benet, L. Z. 1995. Overlappingsubstrate specificities and tissue distribution of cytochromeP450, 3A and P-glycoprotein: implications for drug deliveryand activity in cancer chemotherapy. Mol. Carcinog. 13,129–134.

Yang, J. M., Sullivan, G. F. & Hiat, W. N. 1997. Regulation of thefunction of P-glycoprotein by epidermal growth factorthrough phospholipase C. Biochem. Pharmacol. 53,1597–1604.

Yu, D. K. 1999. The contribution of P-glycoprotein to pharmaco-kinetic drug–drug interactions. J. Clin. Pharmacol. 39,1203–1211.

Zinchuk, V. S., Okada, T., Akimaru, K. & Seguchi, H. 2002.Asynchronous expression and colocalization of Bsep andMrp2 during development of rat liver. Am. J. Physiol.Gastrointest. Liver Physiol. 282, G540–G548.