Transgenic Mice Aberrantly Expressing Human Ornithine

0 downloads 0 Views 4MB Size Report
0 1991 by The American Society for Biochemistry and Molecular Biology, Inc. VOl. 266, No. .... (Sultan) cells overproducing ornithine decarboxylase owing to gene ..... Ackrwwkdgment-The skillful technical assistance of Anne Karp- pinen is ...
THEJOURNAL OF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc.

VOl. 266, No. 29, Issue of October 15, PP. 19746-19751,1991 Printed in U.S. A.

Transgenic Mice Aberrantly ExpressingHuman Ornithine Decarboxylase Gene* (Received for publication, April 1, 1991)

Maria HalmekytoS, Juha-Matti HyttinenS, Riitta SinervirtaS,Merja UtriainenS, Sanna Myohanen$, Hanna-Marja VoipioS, Jarmo WahlforsS, Stina Syrjaneng, Kari Syrjiinenp, Leena Alhonenl, and Juhani JanneSV From the Departments of $Biochemistry& Biotechnologyand §Pathology, University of Kuopio, SF-70211 Kuopio, Finland

We have generated transgenicmice carrying human sion occurs, not only post-transcriptionally, but even at some ornithine decarboxylase gene. Two different transgene post-translational level (2), the short-half life of the enzyme constructs were used: (i) a 5”truncated human orni- apparently having a major impact on its expression (2, 3). In thine decarboxylase gene and (ii) an intact human or- any event, ornithine decarboxylase is one of the most inducnithine decarboxylase gene. Transgenic mice carrying ible (the term is used without any mechanistic implications) the 5“truncated gene did not express human ornithine mammalian enzymes (1). decarboxylase-specific mRNA. Transgenic mice carWe have been especially interested in humanornithine rying the intacthuman ornithine decarboxylase gene decarboxylase. In man, there aretwo ornithine decarboxylaseexpressed human-specific ornithine decarboxylase related sequences. One, actively expressed and amplifiable, is mRNA in all tissues studied. However, as indicated by actual enzyme assays, theexpression pattern was located at chromosome 2 (4) and a processed pseudogene at highly unusual. In comparison with their wild-type chromosome 7 (5). The structure of the enzyme is highly littermates, the transgenicmice exhibited greatly ele- conserved with overall identity at theamino acid level of more vated enzyme activity in almost every tissue studied. than 90% between rodents and human (6-11). The promoter Ornithine decarboxylase activity was moderately ele- regions of murine and human ornithine decarboxylase likevated in parenchymal organssuch as liver, kidney, and wise resemble each other (5,6,12-14). In order to produce animals possibly overexpressing ornispleen. Tissues like heart,muscle, lung, thymus, testis, and braindisplayed an enzyme activity that was20 to thine decarboxylase in their tissues and to study the regula80 times higher than that in the respective tissues of tion of the gene expression of this enzyme, we have generated nontransgenic animals. The offspringof the first trans-transgenic mice carrying the human ornithine decarboxylase gene. At the age of 8 to 9 weeks, the offspring of these mice genic male founder animal did not show any overt abnormalities, yet their reproductive performance was appeared otherwise normal, but showed a strikingly elevated reduced. The second transgenic founder animal,show- ornithine decarboxylase activity in almost all their tissues. ing similaraberrant expression of ornithine decarbox- The increased enzyme activity was in all likelihood a result ylase in all tissues studied, including an extremely high of the expression of human ornithine decarboxylase as huactivity in testis, wasfound to be infertile. Histological man-specific mRNA for ornithine decarboxylase was found examination of the tissues of the latter animal revealed in all tissues studied. The very high ornithine decarboxylase marked changes in testicular morphology. The geractivity in the testis associated with a strikingly elevated minal epithelium was hypoplastic, and the spermato- putrescine content may be a major cause for morphological genesis was virtually totally shut off. Similar exami- and functional alterations found in thistissue. nation of male members of the first transgenicmouse line revealed comparable, yet less severe, histological MATERIALS AND METHODS changes in testis.

Ornithine decarboxylase (EC 4.1.1.17) is a growth-related enzyme, the structure of which is highly conserved among mammalian species (1).The enzyme probably belongs to the “housekeeping”enzymes with a complex pattern of regulation (1).Although the promoter of ornithine decarboxylase gene is very strong and directsits expression to almost all tissues, the gene is expressed inefficiently, and ornithine decarboxylase is a low abundance protein by any definition (1). It is likely that animportant part of the regulation of gene expres* This work was financially supported by the Academy of Finland and by the Finnish Foundation for Cancer Research. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18U.S.C. Section 1734 solelyto indicate this fact. ll To whom correspondence and reprint requests should be addressed Tel: 358-71-163049;Fax: 358-71-211510.

Production of Transgenic Mice-Fertilized oocytes were obtained from superovulated BALB/c X DBA/2 (CD2F1) females mated with CD2Fl males. Pronuclear microinjections were performed using an Eppendorf micromanipulator essentially as described in Ref. 15. The injection volume was about 2 pl and the concentration of the gene constructs 2-5 pg/ml. The number of gene copies injected was about 500. The microinjected zygotes were allowedto reach 2-cell stage and were then transferred into oviducts of pseudopregnant foster females. Gene Constructs Used-The human ornithine decarboxylase gene constructs used for microinjections are shown in Fig. 1.The ornithine decarboxylase genewas originally isolated from human myeloma (Sultan) cells overproducing ornithine decarboxylase owing to gene amplification (16). The intact gene (A, in Fig. 1)represents a BamHIfragment from a genomic X clone later subcloned into pBR328 (17). The insert was cleaved from the plasmid with BamHI and purified with the Geneclean kit (Bio 101 Inc., La Jolla, CA) before injections. This construct was shown to contain all 12 exons and 11 introns together with about 800 and 1000 nucleotides of the 5‘- and 3’flanking regions, respectively (5). The 5”truncated construct (B, in Fig. 1)was obtained by cleaving the intactgene with ThaI restriction endonuclease. This construct was similarly purified with the Geneclean kit before microinjections. Detection of the Transgenes-Tail samples were taken from the

19746

Ornithine Decarboxylase Expression in Transgenic Mice pups at the age of 3 weeks. Tissue DNA was isolated either by the salt precipitation method (18) or by the method of Blin and Stafford (19). Primary detection of the transgenes was carried out with the polymerase chain reaction. Two humanornithine decarboxylasespecific primers were designed to recognize sequences at the third and fourthintron; the amplified fragment (554 nucleotides) thus contained the whole fourth exon (Fig. 1). The two oligonucleotides (30- and 27-mer), corresponding nucleotides 3961 to 3990 (5' primer: 5'-GGCTTACATGTCTTGTTATGGAATGTAGAA-3'),and 4488 to 4514 (3' primer: 5'-GCTATCCATATGTGGCTTAACACGTGG3') of the human ornithine decarboxylase gene, were synthesized with a 381A DNA synthesizer (Applied Biosystems Ltd., Warrington, Cheshire, UK) and purified with the aid of an oligonucleotide purification cartridge (Applied Biosystems). The polymerase chain reactions were carried out in a Hybaid thermal reactor (Hybaid Ltd., Teddington, Middlesex, UK). The reaction was started a t 96 "C for 3 min followed by 36 cycles consisting of 1min a t 95 "C, 1min at 58 "C, and 1.5 min a t 72 "C. One pg of DNA was used as thetemplate. The other ingredients of the reaction were: 10 mM Tris-HC1, pH 9.0, 50 mM KCl, 2 mM Mg& 0.01% gelatin, 0.1% Triton x-100, 0.2 mM dNTP (Pharmacia LKB Biotechnology Inc.) and 1.25 units of Taq polymerase (Promega Corp., Madison, WI). After the completion of the cycles, the samples were left for 5 min at 72 "C. A 15-pl aliquot of the reaction mixture was electrophoresed on a 1.5% agarose gel, stained with ethidium bromide, and visualized with ultraviolet light. The presence of the transgene was confirmed with restriction enzyme analysis. Genomic DNA (10 pg) was digested with H i d 1 1 and MspI restriction enzymes, electrophoresed, blotted (20), probed with the plasmid pODClO/SH (7), and autoradiographed. HindIII and MspI produce human specific restriction fragments (4, 21). Detection of Human and Mouse-specific Ornithine Decarboxylase mRNA in Mouse Tissues-Total RNA was isolated from about 50 mg of tissue using the guanidinium thiocyanate method (22), and the RNA was dissolved in 50 pl of RNasin/water (1 unit/pl; Promega) prior to cDNA synthesis. Any contaminating DNA was removed by DNase treatment. The latter treatment was necessary because mouse genome contains many pseudogenes for ornithine decarboxylase (8, 9). The DNase digestion was carried out in a total volume of 100 pl containing 10 plof 10 X polymerase chain reaction buffer, 2 pl of RNasin (40 units/pl), 10 ml of RQ1 RNase-free DNase (1 unit/pll; Promega), 20 pl of total RNA, and 58 pl of sterile water. The digestion was carried out a t 37 "C for 30 min. After the digestion, RNA was extracted with phenol/chloroform/isoamyl alcohol (25:24:1), precipitated with ethanol, dissolved in water containing RNasin (1unit/pl), and used for cDNA synthesis. The DNase treatment was not necessary when human-specific primers were used in the polymerase chain reaction. The cDNA synthesis was carried out in a total volume of 20 pl containing 2 pl of 10 X polymerase chain reaction buffer, 0.5 pl of RNasin (40 units/pl), 2 pl of dNTP (10 mM; Pharmacia), 1 pl of oligo(T)ls (20 pmollpl), 2 pl of avian myeloblastosis virus reverse transcriptase (5 units/pl; Promega), and 12.5 pl (1pg) of total RNA. Immediately after mixing RNA and the reagents, an aliquot of 10 p1 was taken from the reaction mixture to be used (after denaturation for 5 min in boiling water bath) asnegative control in thepolymerase chain reaction. The cDNA synthesis was carried out at 42 "C, for 60 min in tubes covered with light paraffin oil. After the incubation, an aliquot of 2 pl was taken for the subsequent polymerase chain reaction. For the detection of human-specific ornithine decarboxylase mRNA, the following primers were used in the polymerase chain recreaction. The 5' primer (5'-CCTTCGTGCAGGCAATCTCT-3') ognized human mRNA nucleotides 707 to 726, i.e. a sequence in the middle of exon 7. The 3' primer (5'-GCTGCATGAGTTCCCACGCA3') recognized nucleotides 1323to 1342, i.e. a sequence at thejunction of exons 10 and 11 (6 nucleotides within the exon 10 and 14 nucleotides within the exon 11).Thus, the design of the 3' primer exludes any possibility of genomic DNA being amplified. The primers for murine mRNA detection were designed as follows. The 5' primer (5'AGCACGCCGGCTCTGACGAT-3') recognized nucleotides 1633 to 1652 of murine ornithine decarboxylase mRNA (22) at thejunction of exons 9 and 10. The 3' primer (5"AGACATGGGCAGCGTGCCAT-3') recognized nucleotides 2038 to 2057 in exon 12. The polymerase chain reaction was carried out essentially as described under Detection of the Transgenes (see above) except that only 31 cycles were run. Assay of Ornithine DecarboxylaseActivity and Determination of the Polyamines-Ornithine decarboxylase activity was determined as

19747

described in Ref. 23. Polyamine concentrations were determined with a Hewlet Packard H P 1090 liquid chromatograph. For statistical analyses, Student's t test was used. RESULTS

Generation of Mouse Lines Carrying Human Ornithine Decarboxylase Genes-Initial microinjections were made using intact human ornithine decarboxylase gene (construct A in Fig. 1). However, as almost 100 pups were born from the microinjected zygotes without a single human ornithine decarboxylase-positive animal, we thought that anactive human ornithine decarboxylase genemay not be compatible with mouse embryogenesis. This view was further strengthened by the observations indicating that the litter size tended to be smaller in those animalscarrying embryos microinjected with the active gene. To further test this hypothesis, we started parallel microinjections using the 5"truncated gene (construct B in Fig. I), which did not contain any promoter region and hence in all likelihood would remain transcriptionally silent. It turned out, indeed, that it was much easier to obtain transgenic mice with this construct, as 1 out of 40 pups was transgenic. Finally, however, also 1 male out of 117 pups derived from embryos microinjected with the active ornithine decarboxylase gene was found to be positive. The hypothesis whether active human ornithine decarboxylase gene is not compatible with mouse embryogenesis has not been followed further. The fact, however, remains that out of all the gene constructs (at least 7 different) we have used, the frequency of transgenic pups has been by far the lowest with active human ornithine decarboxylase gene. Fig. 2 shows the detection of the human transgenewith the polymerase chain reaction. The first lane (M) indicates that mouse genomic DNA does not give any signal whatsoever. C represents control in which the microinjected construct (active gene) was used as the template. K1 is a mouse carrying the promoterless human ornithine decarboxylase gene and K2 a mouse carrying the active human ornithine decarboxylase gene. Lanes 1 through 8 are F1 descendants of the K2 founder animal. As shown in Fig. 2,4 out of the 8 pups were transgenic (numbers 1, 4, 6, and 7). Thus, it seems obvious that the founder animal carried human ornithine decarboxylase gene that was inherited in the Mendelian fashion (Fig. 2). The presence of the transgene in the K2 founder animal was confirmed by Southern blotting. Genomic DNA sample obtained from the K2 mouse was digested with MspI and HindIII restriction enzymes, size-fractionated, blotted, and probed with human cDNA-containing plasmid. As shown in Fig. 3, digestion of DNA obtained from human Sultan myeloma (lane S ) with MspI yielded a prominent signal of 2.3-

sy nt

FIG. 1. Gene constructs used for the microinjections. A, human ornithine decarboxylase gene with exon-intron structure. The amplifiable fragment used for the detection of the transgene by the polymerase chain reaction is shown together with the approximate location of the primers. The gene is not drawn strictly to the scale. B, the 5"truncated construct. Other details as for A . nt, nucleotides.

Ornithine Decarboxylase Expression in Transgenic Mice

19748

FIG. 2. Detection of human ornithine decarboxylase transmw, molecular size gene by the polymerase chain reaction. markers (2176, 1766, 1230, 1033, 653, 517, 453, 394, 298, 234, 220, and 154 base pairs). M,mouse DNA;C, positive control (the construct A in Fig. 1);K I , transgenic founder male carrying the 5”truncated humanornithine decarboxylase gene (construct B in Fig. 1); K2, transgenic founder male carrying the intact human ornithine decarboxylase gene (construct A in Fig. 1); I through 8, first generation descendants of the founder male K2. Hind 111

Msp I

M ~

Tg -

S

M

Tg S

size (kbp)

_”

- 23.1 - 9.4

- 6.6 - 4.3

- 2.3 - 2.0 FIG. 3. Restriction enzyme analysis for the detection of man ornithine decarboxylase transgene. Genomic DNA was isolated from the tail sample of the K2 founder animal (carrying intact human ornithine decarboxylasegene) (Tg), from human Sultan myeloma cells (5’) or, from wild-type mouse (M) restricted with either MspI ([eftpanel) or HindIII (rkht panel), electrophoresed, blotted, and probed with pODClO/ZH. kbp, kilobase pairs.

2.4 kilobase pairs insize as did the digestion of DNA obtained from the K2 founder animal (lane Tg in Fig. 3). This is a human-specific signal (21) not found inmouse DNA (lane M in Fig. 3; the faint signal shown in the mouse lane is larger thanthehuman signal). Similar digestion withHindIII yielded a human-specific signal of 5.1 kilobase pairs in size (lanes S and Tg in Fig. 3) not found inmouse DNA (lane M in Fig. 3). Densitometricscanning of theSouthernblots indicated that about24 copies of human ornithine decarboxylase gene were integrated assuming that Sultan myeloma cells contained onecopy of the gene. Detection of Human-specific Ornithine Decarboxylase mRNA in Tissues of the Transgenic Mice-As the sizes of murine and human ornithine decarboxylase mRNA species are so similar in size (9, 16), it would be difficult to detect human-specific mRNA with conventional methods such as Northern blotting. We therefore used reverse transcriptase for cDNA synthesis using total tissue RNA as the template followed by amplification of the product with the aidof the polymerase chain reaction using human and mouse-specific

primers. The primers were designed as to cover one exonexon junction, thus preventing intron-containing genes being amplified. The system worked well in the case of human ornithine decarboxylase mRNA, but, in the case of mouse mRNA, anyDNA contamination had to be removed by DNase treatment prior to the polymerase chain reaction as mouse genome contains a number of apparently intronless pseudogenes. Altogether, we analyzed 9 tissues for the presence of human and mouse-specific ornithine decarboxylase mRNA species. The analyses were made side-by-side with tissues obtained from transgenic mice (F1 descendants) carrying the 5“truncated promoterless human ornithine decarboxylase gene or from F1 transgenic mice carrying the active gene. As shown in Fig. 4, all the tissues (thymus, heart, lung, liver, spleen, kidney, brain, muscle, andintestine)obtained from mice carrying the active human gene gave a positive signal (lanes 10-18) whereas this signalwas absentintissue samples obtained from mice carryingthe promoterless transgene (lanes 1-9). The analyseswere repeated withsomewhat fewer cycles in the polymerase chain reaction and by using mouse and human-specific primers simultaneously. As shown in Fig. 5, mice with the active gene gave a positivesignal(larger fragment seen in the even-numbered lanes), whereas mice with the truncated gene only gave the mouse-specific signal (small fragment seen in the odd-numbered lanes). Although this procedure does not allow proper quantitation of signal intensities, one may observe interesting differences between different tissues. In tissues like thymus (lane 2), heart (lane 4 ) , brain (lane 14), and in skeletal muscle (lane 16), the human-specific signal appeared to be substantially more intense than the corresponding mouse-specific signal. In the case of liver (lane 8), kidney (lane 12), and small intestine (lane 18),there was no difference. Ornithine Decarboxylase Activities in Tissues of Transgenic and Nontransgenic Mice-We next took the same tissues from a series of transgenicpupsand from theirnontransgenic littermates and determined ornithine decarboxylase activities. Table I depicts ornithinedecarboxylase activities in 9 different tissuesof female mice. With theexception of small inteshutine, the enzyme activity was grossly elevated in tissues obm w l 2 3 4 5 m w 6 7 8 9 101112mw

mw C 131415161718

mw

I

FIG. 4. Detection of human ornithine decarboxylase-spe-

transcriptase/polymerase cific mRNA by combined reverse chain reaction assay in tissues of F1 descendants of the K1 founder animal (truncated human ornithine decarboxylase gene) (lanes 1 through 9)and of t h e K2 founder animal (intact human ornithine decarboxylase gene) (lanes 10 through 18). Human mRNA-specific primers (see “Materials and Methods”)were used in the polymerase chain reaction. The tissues analyzed were as follows: thymus (lanes I and IO), heart (2 and II), lung (3and 12), liver ( 4 and 13), spleen (5 and 14), kidney (6 and 151, brain ( 7 and 16), skeletal muscle (8 and I7),and small intestine (9 and 17). mw, molecular size markers (as in Fig. 2). C, positive control (RNA from human Sultan myeloma).

Ornithine Decarboxylase Expression in Transgenic Mice

19749

K7). In attempts to establish a transgenic mouse line with this founder, we found that the animal was infertile. Therefore, the K7 founder animal was killed for further analyses. Table I1 shows tissue ornithine decarboxylase activities in transgenic males derived from the K2 founder animal in the K7 founder animal andin nontransgenic mice. It is apparent that K2 and K7mice displayed a similar ornithinedecarboxylase expression patternin different tissueswith an enhancemwll 12 13 14 mw15 16 17 18 19 Srnw ment of the enzyme activity in almost all tissues. Not only these male animals resembled each other but the expression pattern wasvery similartothatfoundinthetransgenic females (Table I). One may also notice in Table I1 that testis displayed a very high ornithine decarboxylase activity, either the highest found in any tissues analyzed (Table 11; K2) or next to thekidney activity (Table 11; K7). FIG.5. Simultaneous detection of humanandmouse-speTissue Polyamine Contents of the Transgenic Mice-All the in tissues of a descendant cific ornithine decarboxylase mRNA of the K 1 founder animal (truncated human ornithine decar- tissues used for the determinationof ornithine decarboxylase boxylase gene) ( o d d - n u m b e r e d l a n e s ) and of a descendant of activity were also used for polyamine analyses. The general the K 2 founder animal (intact human ornithine decarboxylase picture were slightly surprising. It appeared that most of the lanes) combined by reverse gene) (even-numbered transcriptase/polymerase chain reaction assay. Both human tissues of the transgenic animalswere able tosomehow comand mouse specific primers (see "Materials and Methods") were used pensate the grossly elevated ornithine decarboxylase activity in the polymerase chain reaction assay. The tissues analyzed were as as the concentrationsof putrescine, spermidine, and spermine follows: thymus (lanes 1 and 2 ) , heart (3 and 4 ) . lung (5 and 6), liver were close to those found in the nontransgenic littermates. ( 7 and 8 ) , spleen (9 and IO),kidney (11 and 12), brain (13 and 1 4 ) , The only consistent change in most tissues was an increase skeletal muscle (15 and 16), small intestine (17 and It?), and testis is indicative of rapid (lane 19; only mouse with truncated construct analyzed). The larger of the spermidine to spermine ratio that fragment (626 nucleotides) is the human-specific fragment, and the growth(3). There were two importantexceptions of this smaller one (425 nucleotides) isthe mouse-specific fragment. The general rule: brain and especially testis were apparently unaempty lanes,at the right side of the numbered lanes, represent polym- ble to compensate the enhanced ornithine decarboxylase acerase chain reaction carried out prior to the synthesis of cDNA using tivity as the concentration of putrescine in testes of transgenic the same tissues (see "Materials and Methods"). mw, molecular size 3 times higher males was 5 to 20 times and that in brain about markers (as in Fig. 2); S , RNA from human Sultan myeloma. than in the respective tissues of their nontransgenic litter mates. TABLEI Phenotypic Changes Associated with the Overproduction of Ornithine decarboxylase activity in tissuesof nontransgenic and Ornithine Decarboxylase in Transgenic Mice-As mentioned transgenic female mice above, the second founder male (K7) was infertile. Repeated Values are means f S.D.; n = 3 mice in each group. attempts to mate him with differentfemales did not produce Nontransgenic Transgenic Tissue pregnancies, yet the animal was able toproduce vaginal plugs pmol C02/.30 minlmg tissue indicating that the accessory male sexual glands were func50.9 f 37.4 104.0 f 13.0 Kidney tioning. After the founder animal was killed, it became obvious Liver 2.3 f 2.1 12.2 f 8.2 that there was a dramatic reduction of sperm count.Moreover, Intestine 11.2 f 16.8 9.5 f 7.3 many of the few mature spermatozoa showed malformations, 3.7 I?r 3.1 21.8 f 6.9" Spleen 3.4 f 1.3 54.6 f 17.0h such asdouble heads, and their motility was severely impaired. Thymus Heart 1.6 f 0.7 25.5 f 3.7' Similar examination of several male descendants of the K2 0.7 f 0.3 12.1 f 4.0* Lung founderanimal(alsocarryingtheintacthumanornithine Muscle 0.6 f 0.1 13.9 f 6.1" decarboxylase gene) also revealed that these animals had a Brain 0.2 f 0.1 10.8 f 1.7' significantly decreased sperm count (60.5 ? 10.1% of that " p< 0.05. found in their nontransgenic littermates; n = 4; p < 0.001) hp e 0.01. and a striking reduction (down to 25% of normal) of the ' p < 0.001. number of motile sperm cells. Malformed sperm cells also tained from transgenic female animals. In tissues like liver, occurred much more frequently than in the controls. Out of kidney, and spleen, ornithine decarboxylase activity was mod- the male accessory sexual organs, the preputial glands were erately (2- to6-fold) increased, but, owing to the ratherlarge already macroscopically abnormal in the transgenic males scatter, these changes were not statistically significant. Highly (both in the membersof the established line as well as in the normally significant increases(16- to 23-fold) were found in heart,lung, K7 founder animal). In the transgenic animals, the skeletal muscle, andthymus. However, themoststriking TABLEI1 difference of ornithine decarboxylase activities between transOrnithine decarboxylase activity in tissuesof male descendants ofK2 genic and nontransgenic mice was in the brain tissue, where mouse, the K7 founder mouse, and nontransgenic male mice the activity of ornithine decarboxylase was nearly 70 times Values are means f S.D.; n = 3 mice in each group. higher in the transgenic animals. Interestingly, if one looks Tissue K2 K7 Nontransgenic back at Fig. 5, there apparently existed a rough correlation pmol COJ30 minlmg tissue between the differences of enzyme activities and of the mRNA signal intensities. 1646 14.9 f 6.0 287 f 64 Testis 109 f 106 3604 159 f 177 In the meantime, further microinjections produced another Kidney 42 0.2 f 0.2 Brain 11.3 f 4.7 transgenic founder male carrying the intact human ornithine 21 14.7 f 1.3 0.7 f 0.7 Lung decarboxylase gene (1 out of 132 pups born; designated as 21 0.2 f 0.3 mw1

2

3

4

mw5

6

7

8

9 10

7

Liver

2.4 f 1.5

19750

Ornithine Decarboxylase Expression in Transgenic Mice

vesicle-like organ looked much more solid and wasbrownish, instead of transparent, in color. Most of the tissues of the K7 founder animal and of its nontransgenic littermate were subjected to histological examination. With the exceptionof testis and preputialgland, the tissues appeared to be morphologically normal, although spleen showed an increase inred and a decrease in white pulp indicative of lymphocytedepletion. However, as would be expected, testis of the transgenic animal (Fig. 6A) showed most dramaticmorphological changes in comparison with the normal male (Fig. 6B). The amount of germinal epithelium ( E in Fig. 6) was greatly reduced, the spermatogenesis was practically totally halted, and the numberof Leydig cells ( L in Fig. 6) was increased. This is in striking contrast to the normally advancing spermatogenesis in the testis of the nontransgenic littermate (Fig. 6B).The morphological changes in the testis of the transgenic founder animal resemble a syndrome causing infertility in man and known as “Sertoli FIG.7. The s t r u c t u r e of preputial gland of a transgenic cell-only syndrome” (24).The nameof the syndrome indicates a n i m a l ( A , the K 7 founder animal) and its nontransgenic that the nursing Sertolicells were the only cells found in the littermate ( B ) . The amount of glandular tissue ( G ) is strikingly increased in the transgenic mouse ( A ) . The glands are clearly sebagerminal epithelium,i.e. maturating spermcells werevirtually ceous in type and comprise bulk of the volume. The epithelium ( E ) totally lacking. of the ducts isconsiderably thicker in the transgenic mouse and has Similar histological examination of the male members of undergone a metaplastic change now resembling a squamous epithethe established transgenic line (carrying active human ornilium. In the nontransgenic animal, the glandular tissue is inconspicthine decarboxylasegene)revealed identical morphological uous, and the ducts now occupy the major area. The epithelium of changes in testis, yetpossibly not so severe as those found in the ducts (arrow) is composed of flattened cellswithno signs of the K7 founder animal. Therelikewise was a partial arrest of squamous cell metaplasia (hematoxylin-eosin staining;bars, 100 pm). spermatogenesis. The reason for the changes found in the preputialgland is As mentioned, the macroscopicoutlook of the preputial maybeproduced by gland of the transgenicmales was highly peculiar. Fig. 7 shows not known at the moment, yet they the morphology of the gland, as revealed by light microscopy, enhanced androgenic activity. in a transgenic male (Fig. 7 A ) and its nontransgenic counter- None of the tissues of the female transgenic mice studied showed overt histological abnormalities. part (Fig. 7B). The glandular (G) tissue was markedly increased in the transgenic male (Fig. 7A) in comparison with DISCUSSION the nontransgenic animal(Fig. 7B). I t is likewise obvious that the epithelium ( E ) of the ducts was considerably thicker in There are several possible reasons for an aberrant expresa metaplastic sion of a transgene. One reason may be the property of the thetransgenicanimalandhadundergone change to resemble squamous epithelium. gene construct used. Although the intact human ornithine decarboxylase gene (the gene construct used in presentexper(25) and transiently (17) iments) is expressed both stably apparently with a normal regulation, in transfected cells it is possible that tissue-specificcis-actingregulatory elements may reside outside of the 5‘- and 3”flanking regions that were included in our gene construct. The existence of some silencer element(s) notincluded in our gene constructs is, in fact, supportedby our own experimental results.’ We recently created another transgenic mouse line carrying mouse ornithine decarboxylase promoter fused with bacterial chloramphenicol acetyltransferase gene. Analysis of different tissues of the transgenic mice for chloramphenicol acetyltransferase activity revealedhighlyectopicexpression of the reporter brain. This gene, i.e. extremely high activity in testis and expression pattern closely resembles that found in transgenic male mice expressing human ornithine decarboxylase gene. The idea of a genetic silenceris also supported by some published experimentalevidence. Kahana and hisco-workers FIG.6. A median power view on the testicular structure of (26) found that, during theamplification of mouse ornithine a transgenic mouse ( A , the K7 f o u n d e r a n i m a l ) a n d its non- decarboxylase gene,the expression rate of the gene was higher transgenic littermate ( B ) .The height of the seminiferous epithe- than would be expected from the gene copy number. They ( A ) . The lium ( E ) is markedlyreduced inthetransgenicanimal (26) found that the amplified gene was rearranged witha epithelium is characterized by dramatic decline of the number of all spermatogenic cells (primary and secondary spermatocytes, sperma- major deletion at the far 5’ flanking region. This deletion (26). tids, and spermatozoa) except spermatogonia.The interstitial Leydig possibly relieved the gene fromnegativemodulation cells ( L )are more abundant and prominent than in the nontransgenicThere is no such informationavailable for the human gene. ( B ) ,the germinal epithelium animal ( B ) .In the nontransgenic animal ( E ) is thicker, and the spermatogenesis is advancing normally. The M. Halmekyto, J.-M., Hyttinen, R. Sinervirta, M. Utriainen, S. Myohanen, H.-M. Voipio, J. Wahlfors, S. Syrjanen, K. Syrjanen, L. Leydig cells ( L ) are less abundant and more inconspicuous than in the transgenic animal (hematoxylin-eosin staining;bars, 100 pm). Alhonen, andJ. Janne, unpublished results.

Ornithine DecurboxyEase Expression in Transgenic Mice Also relevant to our observations are therecent findings of Rauth et al. (27) dealing with murine adenosine deaminase promoter. They found that while endogenous adenosine deaminase activity in brain was extremely low, the situation was just the opposite in transgenicmice. Adenosine deaminase promoter-driven reporter gene showed consistent high level expression in the brain of the transgenic mice. The authors (27) proposed the existence of a brain-specific suppressor sequence normally associated with the murine adenosine deaminase gene, but not included in their gene construct. We found the same phenomenon, i.e. a strikingly enhanced expression of brain ornithine decarboxylase in thetransgenic animals. Unusual expression and altered tissue specificity of a foreign gene may also be modulated by chromosomal position effect in a particular transgenicmouse line as the integration occurs at random chromosomal sites (28). However, this possibility is somewhat unlikely as also the second male founder animal showed almost identical aberrant expression pattern for ornithine decarboxylase. It is remarkable that the tissues were able to compensate the strikingly enhanced ornithine decarboxylase activity with very little changes in the levels of the polyamines. The only change in this respect in most tissues was an elevation of the ratio of spermidine to spermine which is typical to rapidly growing tissues (3). The only exceptions of this rule were testis and brain, tissueswhich in there was a dramaticincrease in putrescine content. The high ornithine decarboxylase activity and enhanced accumulation of putrescine in the brain of the transgenic animals did not apparently result in any morphological changes. It, however, remains to be seen whether the high putrescine content influences functional properties of the brain tissue. Incontrast to brain, high ornithine decarboxylase activity and strikingly increased accumulation of putrescine apparently resulted in major morphological and functional alterations in testicular tissue with grossly impaired spermatogenesis as the major feature. The reasons for these changes are not known at themoment, but it appears that ornithine decarboxylase activity and the accumulation of the polyamines are closely connected to the testicular development. Interestingly, sexually mature rodent testis seems to be one of the most active tissues expressing ornithine decarboxylase (29). The fluctuations of ornithine decarboxylase expression likewise seem to be finely timed duringthe development (30, 31). Thus,the constitutively overexpressed ornithine decarboxylase and theresulting continuous high levels of putrescine may interfere with the testicular development. One may also notice that thenumber of Leydig cells, producing testosterone, was increased, and the preputial gland displayed morphological changes indicative of enhanced androgenic activity. It is thus possible that the situation has likewise created a major hormonal imbalance. Irrespective of the molecular mechanisms responsible for the aberrant expression of human ornithine decarboxylase gene.

19751

in transgenic mice, the fact remains that we now have an ornithine decarboxylase “overproducer” mouse line. This line is valuable for a further assessment of the relationship between an enhanced ornithine decarboxylase activity and the metabolism of a given tissue. These mice may be especially suitable for carcinogenesis studies. Exceedingly interesting, in this respect, is the view, supported in fact by some sets of experimental evidence (32, 33), that an overproduction of ornithine decarboxylase would confer a growth advantage to mammalian cells. Ackrwwkdgment-The skillful technical assistance of Anne Karppinen is gratefully acknowledged. REFERENCES 1. Hayashi, %-I. (ed) (1989) Ornithine Decarboxylase:Biology,Enzymology, and Molecular Genetics, pp. 1-147, Per amon Press, New York 2. van Daalen Wetters, T., Brabant, M., an8 Coffino, P. (1989) Nucleic Acids Res. 17,9843-9860 3. Janne, J.,Poso, H., and Raina, A. (1978) Biochim. Biophys. Acta473,241293 4. Whqvist, R., Makela, T. P., Sep h e n , P., Jiinne, 0.A., Alhonen-Hongisto, L.. J h n e . J.. Grzeschik. K . - k and Alitalo. K. (1986) Cvtonenet.Cell . &net. 42; 133-140 ’ 5. Hickok, N. J., Wahlfors, J., Crozat, A., Halmekyto, M., Alhonen, L., Janne, J., and J h n e , 0.A. (1990) Gene (Amst.) 93,257-263 6. Fiz erald M C and Flanagan, M. A. (1989) DNA (NY) 8.623-634 7. Hictok N . J depphen P. J., Gunsalus, G. L., and Janne, 0.A. (1987) DNA’(NYy6, 179-18j 8. Kahana, C., and Nathans, D. (1984) Proc. Natl. Acad.Sci. U.S. A. 8 1 , 3645-3649 9. Kontula, K. K., Torkkeli, T. K., Bardin, C. W., and Janne, 0.A. (1984) Proc. Natl. Acad. Sci. U. S. A. 81,731-735 10. van Steeg, H., van Oostrom, C.,T. M., van Kreyl, J. W. M., Schepens, J., and Wlenng,B. (1989) Nuclele Accds Res. 17,8855-8856 11. Wen, L., Huang. -. J.-K.. and Blackshear, P. J. (1989) J. Biol. Chem. 264. 9016-9021 12. Brabant, M., McConlogue,L., van Daalen Wetters, T., and Coffino, P. (1988) Proc. Natl. Acad. Sci. U. S. A. 86,2200-2204 13. Coffino, P., and Chen E. L.(1988) Nucleic Acids Res. 16,2731 14. Eisenber L M and’Janne, 0.A. (1989) NucleicAcids Res. 17,2359 15. Ho an, 8 , Con&ntini, F., and Lacy E. (1986) Mani ulating the Mouse lfmbryo, pp. 1-322, Cold Spring HarborLaboratory, &Id Spring Harbor, I

.



I

NV I .

I

16. Leinonen, P., Alhonen-Hongisto, L.,Laine, R., J h n e , 0.A., and Janne,J. (1987) Bmhem. J. 242,199-203 17. Halmekp, M., Hirvonen, A., Wahlfors, J., Alhonen, L., and Janne, J. (1989 BLochem.Bio h s Res Commun. 162,52&534 18. Miller, S. A., Dykes, k , and Polesky, H. F. (1988) Nucleic Acids Res. 16,1215 19. Blin, N., and Stafford, D. W. (1976) Nwleic Acids Res. 3,2303-2308 20. Southern, E. M. (1975) J. Mol. Biol. 98,503-517 21. Alhonen-Hon ‘sto, L., Leinonen, P., Sinervirta, R Laine, R., Winqvist R., Alitalo, Janne, 0. A., and Janne,J. (1987)kochem. J. 242,2051

8.

21 n

E.,

22. Chomcz ski, P.,,and Sacchi, N. (1987) A d . Biochqn. 162,156-159 23. Janne, r a n d Williams-Ashman, H. G. (1971) J. Bwl. Chem. 246,17251732 24. Mustofi F. K., and Davis, J. D Jr. (1990) in Anderson’s Pathology(Kissane, J. M.,’ed) 9th ed, pp. 871-9i7 C. V. Mosby Co., St. Louis 25. Holttii, E., Hirvonen, A., Wahlfbrs, J., Alhonen, L., Janne, J., and Kallio, A. (1989) Gene (Amst.) 83,125-135 26. Katz, A. and Kahana, C.(1989) EMBO J.8,1163-1167 27. Rauth, 8., Yang K . 4 . Selbold, A.M., Ingolia, D.E., Ross, S. R., and Yeung, C.-Y. (1990) somat. Cell. Mol. Genet. 2, 129-141 28. Al-Shawi, R.,Kinnaird, J., Burke, J., and Bishop, J. 0.(1990) Mol. Cell. Biol. 10,1192-1198 29. Blackshear, P. J., Manzella, J. M., Stumpo D. J., Wen, L. Huang, J.-K., Oyen, O., and Young, W. S . , III (1989)M h Endocrin@ 9,138-78 30. Kaiia A. Toppan, J. Mall, P. Kangasnleml M., Alclvar A. A Hecht, ’fi B., aAd Parvinen:M. (199d) Mol. Ceil. Enhocrinol. 73,)45-52’ 31. Shubhada, S., Lin, S.-N., Qian, Z.-Y., Steinberger, A., and Tsai, Y.-H. (1989) J. Androl. 10.45-151 32. Alhonen-Hon ‘sto, L. Kallio A,, Sinervirta R Janne 0.A., Gahmberg, C. G., and &ne, J: (1985) Eiochem. J. 2 2 9 , ’ b 7 1 ~ 33. Polvinen, K., Sinervirta, R., Alhonen, L., and Janne, J. (1988) Biochem. Biophys. Res. Commun. 166,373-378