Proc. Natl Acad. Sci. USAVol. 78, No. 10, pp. 6376-6380, October 1981Genetics

Microinjection of a rabbit fi-globin gene into zygotes and itssubsequent expression in adult mice and their offspring

(gene transfer/interspecific gene expression)

THOMAS E. WAGNER*, PETER C. HOPPEt, JOSEPH D. JOLLICK*, DAVID R. SCHOLL*4,RICHARD L. HODINKA*, AND JANICE B.-GAULT**Deparnefnts of Chemistry, Zoology-Microbiology, and Biomedical Sciences, Ohio University, Athens, Ohio.45701; and tThe Jackson Laboratory,Bar Harbor, Maine 04609

Communicated by Elizabeth S. Russell, July 9, 1981

ABSTRACT We have transferred a gene coding for rabbit (3-globin into the male pronucleus of mouse zygotes by direct mi-croinjection. Some of these zygotes developed into mature micewhich contained this gene and appeared to be producing a rabbitglobin. Evidence for the presence ofthe gene in these animals wasprovided by Southern blot hybridization analysis. Evidence for theexpression of the rabbit gene in these transformed mice and theiroffspring was provided by hemoglobin isoelectric focusing analysisand specific serological reactivity between mouse anti-rabbithemoglobin antiserum and a hemolysate from the mice that de-veloped from the microinjected zygotes. The use of this zygotetransformation may allow the introduction and expression of abroad range of genetic elements in mammals.

The transfer of specific functional genes from one mammalianspecies to another thus far has been realized only in cell culturesystems. Cloned rabbit ,3globin gene sequences have been sta-bly introduced into thymidine kinase (TK)-deficient mutantmouse cells by DNA-mediated cotransformation using the rab-bit f-globin gene and herpesvirus TK gene (tk) sequences (1).This approach has been extended to use the cellular TK (2),adenine phosphoribosyl transferase (3), and hypoxanthine phos-phoribosyltransferase (4, 5) genes as unlinked but selectablemarkers. Use of a mutant hamster gene coding for an altereddihydrofolate reductase as a selectable marker offering meth-otrexate resistance allows the introduction and amplification ofa broad range of genetic elements into various cell lines (6).Although these elegant cotransformation experiments makepossible direct gene transfer between mammalian species, theyare restricted to cell culture systems because only 1 cell in105-107 is transformed, and these cells must be selected for inrestrictive media.

Incorporation of specific genes into the genome of mam-malian embryos would provide a useful in vivo system for anal-ysis of the control of gene expression during differentiation.Because the low transformation efficiency of direct cotransfor-mation makes this method unsuitable for embryo transforma-tion,, alternative methods are required. Recently, high-effi-ciency transformation of cultured mammalian cells has beenaccomplished by direct microinjection of specific DNA se-quences into the cell nucleus (7, 8), suggesting that this methodof cell transformation may be effective with embryos or fertil-ized eggs. Jaenisch and Mintz (9) using whole simian virus 40and Gordon et aL (10) using a recombinant plasmid composedof segments of herpes simplex virus, simian virus 40, and bac-terial plasmid pBR322 have provided evidence that DNA mi-croinjected into mouse embryos may be found in the resultantoffspring. The probability of formation of stably transformedmammalian embryos as a result of direct gene microinjection

may depend on the stage ofdevelopment of the egg or embryoat the time of injection. The optimal time for gene microinjec-tion may coincide with the fertilization process.

Fertilization is a complex, multistep phenomenon initiatedby the interaction and fusion of the spermatozoon and egg andculminated by the association of the two groups of chromo-somes, one derived from the maternal pronucleus and the otherderived from the paternal pronucleus. During the early stagesof the fertilization process, just after the spermatozoon pene-trates into the egg, the sperm nucleus undergoes a complexchromosomal decondensation process (11-13) after which themale chromosomal complement appears to undergo changes inchromosomal composition prior to merging with the female pro-nucleus (14). Although little is known about this stage of fertil-ization in the mammal, studies of the sea urchin suggest thatsignificant changes within the male pronucleus and chromatinare brought about by proteins emanating from the oocyte cy-toplasm and female pronucleus (15, 16). Microinjection of spe-cific DNA sequences into the male pronucleus during pronu-clear "processing" ofthe male chromosomes might result in theappropriate delivery ofthese sequences into the zygote nucleusalong with the male chromosomal complement.On the basis of these considerations, an estimated 20,000

copies ofeither purified rabbit (globin gene fragment or a rab-bit (3-globin gene-containing plasmid were microinjected intomale pronuclei of mouse zygotes. The resulting embryos werecultured in vitro to morulae or blastocysts and transferred intopseudopregnant foster mothers. We report here not only thepresence of rabbit (3-globin gene in offspring developed fromthese microinjected mouse zygotes but also serological and elec-trophoretic evidence for the production of a rabbit globin pro-tein in these mice.

MATERIALS AND METHODSCollection of Zygotes. Fertilized.eggs at the pronuclear stage

(the male and female pronuclei being separated and distinguish-able within the cytoplasm) were collected from the oviducts ofC57BL/6J females that had been mated to LT/Sv males. Afterremoval of the surrounding cumulus cells in culture medium(17) with bovine testis hyaluronidase (1 mg/ml), pooled zygotesfrom several females were washed in fresh medium and stored,until micromanipulation, in a depression slide containing cul-ture medium overlayered with paraffin oil in an atmosphere of5% COJ5% 02/90% N2 at 3rC.

Micromanipulation. Injection pipets (external diameter, 1-2g&m) and holding pipets (60-70 gm) were prepared from Pyrextubing (1.2 mm; Corning) as described (18). Manipulation ofthe

Abbreviations: TK, thymidine kinase; PINaCl, 0.15 M NaCI/0.01 Msodium phosphate, pH 7.2* Present address: Roche Institute of Molecular Biology, Nutley, NJ07110.

6376

The publication costs ofthis article were defrayed in part by page chargepayment This article must therefore be hereby marked "advertise-ment" in accordance with 18 U. S. C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

July

28,

202

0

Genetics: Wagner et aL

pipets for holding and injection of the zygote was accomplishedby using Leitz micromanipulators and paraffin oil-filled Ham-ilton syringes (TP 1750 LT). A small drop of culture mediumwith five or six zygotes and ofrabbit [globin chromosomal gene

solution were placed on a special microscope slide and coveredwith paraffin oil. About 10 pl of either purified rabbit -globinfragment or Z-pCRI/RchrpG-1 plasmid (approximately 20,000DNA molecules) was drawn into the injection pipet which was

then moved to the drop containing the eggs. A zygote was po-

sitioned onto the holding pipet so that the male pronucleus wasin juxtaposition to the injection pipet for subsequent injectionof the 3globin gene solution into the pronucleus (Fig. 1). Aftermicroinjection into all the zygotes, they were removed from thedrop of medium on the microscope slide and placed in culturetubes (17) for preimplantation development during 5 days ofculture. Eggs developing to morulae or blastocysts were trans-planted into uteri of B6SJLFI hybrid foster mothers (day 3 ofpseudopregnancy) and were carried to term.

Preparation of Rabbit 1-Globin Gene Fragment. Esche-richia coli HB101 containing the hybrid plasmid Z-pCRI/Rchr/3G-1 was kindly provided by R. Flavell (19). The plasmidwas extracted by using a lysozyme/Triton X 100/EDTA lysisprocedure and purified by CsCVethidium bromide density gra-dient centrifugation followed by agarose gel electrophoresis inlow-melting-point agarose. The purified plasmid which con-

tained the rabbit f3-globin gene was then digested with Hha I[plasmid pCR1 has numerous Hha I sites whereas the 3-globingene has none (19)]. The 6200-base-pair 6globin gene DNAfragment (19) was then purified by electrophoresis of the HhaI digest in low-melting-point agarose. The ,3-globin fragmentDNA band was extracted from the agarose by melting the gelat 650C and extracting with phenol. The (-globin gene frag-ments contained approximately 50 ,ug of DNA per ml.Hb Preparation. Hb was prepared as a purified hemolysate

from New Zealand White rabbits, control B6LTF1 hybrid mice,and B6LTF1 hybrid mice developed from zygotes microinjectedwith rabbit ,-globin gene DNA sequences. Erythrocytes were

washed exhaustively with phosphate-buffered saline [0.15 MNaCl/0.01 M sodium phosphate, pH 7.2 (PJ/NaCl)], lysed indistilled water, and centrifuged at 12,000 x g for 10 min to re-

move the erythrocyte membrane. Prior to application ofthe Hbto agarose immunodiffusion plates, the solutions were filteredthrough a 0.45-,um filter and adjusted to a concentration of 100mg/ml.

Rabbit Globin. Rabbit globin was prepared from an eryth-rocyte hemolysate by acetone extraction ofthe heme-iron com-

FIG. 1. Microinjection into the mouse zygote male pronucleus.(x 1200)

Proc. Natl. Acad. Sci. USA 78 (1981) 6377

Table 1. Pre- and postimplantation development of B6LTF1hybrid zygotes after microinjection of rabbit t3-globinchromosomal gene into male pronucleus

Fostermothers, no.

Embryos Off-Eggs to foster spring

DNA Zygotes, cleav- mothers, Preg- born,injected no. ing, no. no. Total nant no.

Hha Ifragment 143 139 120 12 8 28

Plasmid* 169 152 91 9 3 18* Z-pCRI/RchrfG-1

plex (20). Packed, washed, rabbit erythrocytes were lysed in 2ml of distilled water, and the resulting hemolysate was drippedslowly into a stirred solution of 60 ml of acetone and 1.2 ml ofconcentrated HC1 at -70TC. The resulting precipitated rabbitglobin was allowed to stand for 15 min at - 70TC in the extractionmixture; then it was washed twice with acetone at -20TC anddried under reduced pressure at liquid nitrogen temperatures.The dried precipitate was dissolved in 2 ml ofdistilled H20 anddialyzed exhaustively against PJNaCl.

F1 Hybrid Mouse Anti-Rabbit Hb Antiserum. Mouse anti-rabbit Hb antiserum was raised in B6LTF1 hybrid mice. Eachanimal received a500 Ag priming dose (subcutaneously, dividedamong four sites) of rabbit Hb in distilled water emulsified withan equal volume of complete Freund's adjuvant. After 1 and 2weeks, all animals received second and third injections of 10 ,ugof rabbit Hb emulsified in incomplete Freund's adjuvant ad-ministered subcutaneously and intraperitoneally, respectively.After 4 weeks, each animal was injected intraperitoneally with600 ,Ag of rabbit Hb in complete Freund's adjuvant. The im-munized animals were bled after 5 weeks and the serum wasseparated, tested against rabbit Hb, and preserved in the pres-ence of sodium azide.

Gel-Immobilized Rabbit Hb. Rabbit Hb (30 mg) was coupledto Sepharose 4B beads by reaction with 1 g of washed, CNBr-activated Sepharose 4B (Pharmacia) in 0.1 M NaHCOJ0.5 MNaCl, pH 8.0, for 3 hr at room temperature. Then the beadswere treated with 1 M Tris (pH 8.0) and the Hb-Sepharose gelwas washed exhaustively to remove unbound Hb. The gel-im-mobilized rabbit Hb was used to prepare a miniature columnfor the selective removal of anti-rabbit Hb antibodies from im-munized mouse serum.

Agarose Immunodiffusion. Reactivity of Hb antigens withmouse anti-rabbit Hb antiserum was detected by double-dif-fusion analysis on 1% agarose slides. Antigen wells contained15 Al of each Hb antigen (100 mg/ml); the antibody well con-tained 15 Al of anti-rabbit Hb antiserum. Each well was refilledtwice during the 24-hr incubation at 37°C.

Indirect Immunofluorescence Assay. Smears of washederythrocytes fixed in methanol were incubated with partiallypurified mouse anti-rabbit Hb antiserum (from which the al-bumin fraction had been removed by precipitation with octanolat pH 5.5-6.0) for 30 min at 37°C, washed in PJNaCl, andtreated with mouse tissue-absorbed, fluorescein-labeled, goatanti-mouse gamma globulin (Antibodies, Inc., Davis, Califor-nia). The immunofluorescent stained cells were then washedin PJNaCl for 15 min and observed by UV fluoresencemicroscopy.

Isoelectric Focusing. Saline-washed blood cells were lysedand the hemolysate was prepared for focusing as described byWhitney et aL (21). Focusing was performed in 6 x 120mm tubegels containing 1% zero-Mr agarose (Bio-Rad) and 2% Ampho-

-------

-j:..

i. -. .!AA,sow.

Dow

nloa

ded

by g

uest

on

July

28,

202

0

Proc. Natd Acad. Sci. USA 78 (1981)

FIG. 2. Ouchterlony agarose geldiffusion reactions. (Left) Erythrocytesof mice from injected zygotes containa molecular species that reacts specif-ically with antiserum produced againstrabbit Hb. Wells: 1, normal B6LTF1hybrid Hb preparation; 2, rabbit Hb; 3,Hb from mice developed from the in-jected zygotes; A, anti-rabbit Hb anti-serum; B, normal nonimmunized mouseserum. (Right) Identity between one ofthe dissociated rabbit globin chainsand the molecular species present inerythrocytes of mice developed frommicroinjected zygotes. Wells: 1, 4, and6, rabbit Hb preparation; 5, normalB6LTF1 hybrid Hb preparation; 2, dis-sociated rabbit globin chains; 3, Hbpreparation from mice developed frommicroinjected zygotes; A, anti-rabbitHb antiserum.

lines (pH 5-8) (Pharmacia). The cathode and anode buffers were0.02 M NaOH and 0.01 M H3PO4, respectively. Samples were

run at constant power (0.83 W) for 150 V hr. Gels were observedeither without staining or after staining with Coomassie brilliantblue R-250 and destaining.

Hematology. Erythrocyte counts, hematocrit, and hemoglo-bin values were determined by using a Coulter Counter ZBI,(Coulter Electronics). Reticulocyte counts were determinedmicroscopically after staining with methylene blue.

Analysis of Mouse Chromosomal DNA for Rabbit 1..GlobinDNA Sequences. Liver DNA from control B6LTF1 hybrid miceand from a B6LTFj hybrid mouse developed from a zygotemicroinjected with rabbit ,B-globin gene DNA sequences was

isolated by the method of Jeffreys and Flavell (22). For eachanalysis, 30-40 ,ug ofeach DNA was digested with Taq I. Eachdigested sample was separated on a 1% agarose gel by electro-phoresis at 25 mA for 16 hr. The DNA was then transferred toa BA-85 (Schleicher & Schuell) nitrocellulose membrane ac-

cording to the method of Southern (23). Molecular size stan-dards were provided by restriction endonuclease digests of theplasmid Z-pCRI/RchrpG-1 or A DNA.

Rabbit ,&globin probe was prepared from Z-pCRI/Rchrf3G-1 by the method ofSummers (24). Hybridization was performedat 65°C for 48 hr in 0.45 M NaCV0.045 M Na citrate, pH 7,containing 3X SSC 0.2% bovine serum albumin, 0.02% Ficoll,and 0.02% polyvinylpyrrolidone). After hybridization, the fil-ters were washed three times for 4 min each and once for 30min in 0.15 M NaCV0.015 M Na citrate at 37°C. Finally, filterswere washed in 0.03 M NaCV0.003 M Na citrate at 659C for

A B

FIG. 3. Isoelectric focusing ofhemolysates from normal B6LTF1hybrid mice (A) and mice developedfrom microinjected B6LTF1 zygotes(B). The gels were loaded with 0.7mg of hemolysate protein and pho-tographed immediately after com-pletion of focusing. The bands arenot fixed or stained. All bands ob-

+ served were red and were in the re-gion pH 7.2-pH 7.0.

4 hr. Filters were exposed to Kodak X-Omat (XR-7) film at-700C with an intensifier screen for 7 days.

RESULTSZygote Micro'mjection. Data illustrating the pre- and postim-

plantation development of the microinjected zygotes are com-piled in Table 1.

Immunodiffusion Analysis. Washed erythrocytes were pre-pared from blood samples taken from each of the offspring re-sulting from the injected hybrid mouse zygotes and analyzedby an indirect immunofluorescence assay using B6LTF1 hybridmouse anti-rabbit Hb antiserum and fluorescein-labeled goatanti-mouse gamma globulin. Although a significant number ofthe samples showed some positive immunofluorescence com-pared to the nonfluorescent control B6LTF1 hybrid mouseerythrocytes, the immunofluorescence intensity of the individ-ual samples varied. Five samples (nos. 4, 12, 20, 24, and 26)which showed a distinct positive immunofluorescence reactionwere pooled and used for the following immunodiffusion andisoelectric focusing analysis.

Agarose gel immunodiffusion using F1 hybrid mouse anti-rabbit Hb antiserum showed a major band ofprecipitation withthe control rabbit Hb, no evidence of precipitation with Hbprepared from control mice, and a major band of precipitationwith Hb prepared from the mice that developed from thetreated zygotes (Fig. 2 Left). The major precipitation band be-tween the experimental mouse Hb and the mouse anti-rabbitHb antiserum shows a pattern of fusion, indicating serologicalidentity with non-heme-contaning globins prepared from rab-bit Hb (Fig. 2 Right).

In order to establish that the immunoprecipitation bandsbetween the B6LTF1 hybrid mouse anti-rabbit Hb antiserum

Table 2. Hematological values in experimental mice

Erythro- Meancytes, Hemato- cell

no. X 10-6/ Hb, crit, volume, Reticulo-Mouse mm3 g/dl % Am3 cytes, %

4 11.7 17.2 50.0 43.0 3.212 11.0 16.9 47.0 43.0 2.520 11.0 16.5 51.0 43.0 7.524 10.8 16.9 49.0 44.0 5.326 10.7 16.4 48.0 44.0 2.2

Normalcontrol* 10.06 15.24 44.04 43.71 0.5

±0.30 ±0.32 ±2.03 ±1.49

* Normal hybrid, shown as mean ± SD (n = 14).

6378 Genetics: Wagner et al.

Dow

nloa

ded

by g

uest

on

July

28,

202

0

Proc. NatL Acad. Sci. USA 78 (1981) 6379

and the Hb preparation from mice developed from zygotesmicroinjected with rabbit -globin genes were the result of aspecific immunological reaction, a control immunodiffusionanalysis was performed with B6LTF1 hybrid mouse anti-rabbitHb antiserum that had been absorbed by passage through acolumn containing rabbit Hb bound to Sepharose 4B. The ab-sorbed antiserum showed no reactivity with either rabbit Hbor the Hb from the mice from the treated zygotes.

Isoelectric Focusing Analysis. The B6LTF1 hybrid normalmice ofHb type Hbbs-Hbaa-Hbab showed a major Hb band andtwo minor bands cathodal to the major band (pH 7.0-7.2, thepI range for both rabbit and mouse Hb) upon isoelectric focusingof their hemolysates (Fig. 3). Hemolysates from the B6LTF1hybrid mice developed from treated zygotes also showed thesethree Hb species and an additional band anodal to the major Hbband. Overloaded isoelectric focusing gels of normal B6LTF1hybrid mouse hemolysates and hemolysates from mice devel-oped from treated zygotes showed additional bands at andbelow pH 6.8, the approximate pI of free globin chains (Fig. 4)An additional band is present in the experimental B6LTF1 hy-brid mouse hemolysate in this region of the pH gradient.

Hematology. The hemoglobin concentration, erythrocyteconcentration, and hematocrit values for the five experimentalmice were greater than those of normal control B6LTF1 mouseblood (Table 2). Also, the reticulocyte concentration in the ex-perimental mouse blood was increased 5-fold or greater overthe normal B5LTF1 control mouse blood, suggesting a mildthalassemia in the mice developed from zygotes microinjectedwith the rabbit ,B-globin gene.

Progeny Immunodiffusion Analysis. Mice (nos. 24 and 26)that were bled for the pooled immunodiffusion analysis samplewere mated; and no. 24 gave birth to eight offspring. Ofthe sixsurviving offspring (two died shortly after birth), five were sac-rificed and their hemolysates were analyzed against B6LTF1hybrid mouse anti-rabbit Hb antiserum. All five progenyshowed a distinct immunoprecipitation band between their Hb

A

B

FIG. 4. Isoelectric focusing of hemolysates from experimentalmice developed from B6LTF1 zygotes microinjected with rabbit P-glo-bin gene (A) and from normal B6LTF1 hybrid mice (B). Gels wereloaded with 3.1 mg of hemolysate protein. The gel scan shown repre-sents the region of the gel from pH 7.0 to pH 6.5. After focusing, thegels were fixed and stained with Coomassie brilliant blue R-250 andwere scanned at 590 nm. Two bands with a pI in the 6.8 region are seenin the experimental hemolysate; only one band with that pI is seen inthe normal control B6LTF1 hemolysate.

Vi

FIG. 5. Detection ofDNA frag-ments hybridizing to Z-pCRI/RchrPG-1 sequences: 40 pg of liverDNA from microiijected B6LTF1hybrid (lane 2) and from controlB6LTF1 hybrid (lane 1) and 1 ng ofZ-pCRI/RchrBG-1 (lane 3) werecleaved with Taq I (80 units) at65°C for 2 hr, transferred to filters,and hybridized to a 32P-labeled Z-pCRI/Rchr,/G-1 probe. Arrow-head, 2.7 kilobases.

wells and the antiserum well and showed four distinct Hb bandsupon isoelectric focusing.

Analysis of Mouse Chromosomal DNA for Rabbit ,-GlobinDNA Sequence. DNA from the liver ofone ofthe experimentalB6LTF1 hybrid mice (no. 24) was digested with Taq I. Thisdigest was analyzed by Southern hybridization using the plas-mid Z-pCRI/Rchr3G-1 as a 32P-labeled probe. The Taq I digestof the experimental mouse liver DNA showed a fragment, atthe same position as the 2.7-kilobase fragment from a Taq Idigest of Z-pCRI/Rchrf8G-1, hybridizing with the probe (Fig.5). This Taq I fragment arising solely from within the 6-kilobaserabbit genomic insert in Z-pCRI/Rchr1BG-1 (25) substantiatesthe presence of the rabbit ,B3globin gene in the genome of theB6LTF1 hybrid experimental mouse (no. 24) which was sug-gested, by immunodiffusion and Hb isoelectric focusing anal-ysis, to contain rabbit globin protein. The additional smallerfragments hybridizing with the probe arose from other se-quences within the plasmid.

DISCUSSION

New technologies have made possible the isolation and cloningofDNA sequences coding for specific gene products. The stableintroduction of these gene sequences into the genome ofmam-malian zygotes would facilitate the study of gene expressionduring development and in mature animals derived from thesezygotes.

In the present study, we have demonstrated the presence ofrabbit ,B-globin gene sequences in the genome of one of ourexperimental mice developed from zygotes microinjected withthis purified rabbit gene. The presence of a 2.7-kilobase frag-ment from within the 6-kilobase rabbit genomic insert in theZ-pCRI/RchrfG-1 plasmid in the Taq I digest of the experi-mental mouse liver DNA confirms the presence of the rabbitgene in mouse no. 24. The results shown in Fig. 5 do not disclosewhether or not the rabbit gene sequences are integrated intothe mouse chromosome or, if integrated, what the position ofthis integration might be.We have also demonstrated the presence ofan additional Hb

component in the erythrocytes of these experimental mice aswell as the presence of what appears to be an additional globinspecies by isoelectric focusing analysis. That these new molec-ular species in the erythrocytes of the experimental mice mayresult from the synthesis of a rabbit globin protein is suggestedby serological reactivity between experimental mouse hemo-lysates and mouse anti-rabbit Hb antiserum. Evidence for theproduction of a rabbit globin protein in the experimental miceis provided by three observations.

Genetics: Wagner et aL

Dow

nloa

ded

by g

uest

on

July

28,

202

0

Proc. Natd Acad. Sci. USA 78 (1981)

A characteristically red-colored Hb species not present incontrol B6LTF1 hybrid mouse erythrocytes is observed as adistinct band at pH 7.0 in the isoelectric focusing profile of he-molysates from the experimental B6LTF1 hybrid mice (Fig. 3).The observation of another unique component present in theexperimental mouse hemolysates and banding at pH 6.8 nearfree mouse globin chains in the isoelectric focusing profile (Fig.4) suggests that a unique globin species is present in some ofthe experimental mice both as free globin chains and in intactHb molecules. Additionally, a distinct band of precipitation isobserved, during agarose double-diffusion analysis, betweenthe wells containing a Hb preparation from the experimentalmice and the well containing mouse anti-rabbit Hb antiserumwhereas Hb prepared from control F1 hybrid mice showed thecomplete absence of any reaction with the antiserum (Fig. 2Left). This precipitation band also showed complete identitywith a precipitation band observed between a rabbit globinpreparation and the antiserum (Fig. 2 Right). Finally, absorp-tion of the mouse anti-rabbit antiserum with gel-immobilizedrabbit Hb abolishes the reaction between the absorbed anti-serum and Hb prepared from the experimental mice. This ob-servation greatly lessens the possibility that any antibodiesother than anti-rabbit Hb antibodies caused the precipitationreaction with Hb from the experimental mice. This serologicalevidence is strengthened by the genetic strain identity betweenthe mice used to produce anti-rabbit Hb antiserum and theexperimental mice. This identity diminishes the possibility thatspurious immunological crossreactivity may have occurred. Theresults confirm the production ofwhat appear to be at least smallamounts of rabbit globin in the erythrocytes of mice developedfrom zygotes microinjected with rabbit (-globin gene sequences.

Mice developed from zygotes microinjected with the rabbitf-globin gene show increased values for erythrocyte concen-tration, Hb concentration, hematocrit, and reticulocyte con-centration suggestive ofincreased erythropoiesis (Table 2). Thisobservation suggests that these mice may be mildly thalassemicresulting from an overproduction of (-globin chains.One of the mice bled for the immunodiffusion studies (no.

24) and eventually sacrificed in order to obtain the experimentalmouse liver DNA was mated to a male showing Hb immuno-reactivity with anti-rabbit Hb antiserum. All of the offspringfrom this mating that were analyzed also showed Hb immu-noreactivity with the anti-rabbit antiserum and displayed fourHb bands in their isoelectric focusing profiles. This observationsuggests that the rabbit (-globin gene introduced into our ex-perimental mice by direct zygote microinjection may beheritable.

Expression of the rabbit 3-globin gene in mice that devel-oped from zygotes microinjected with DNA sequences con-taining this gene may be dependent upon three important fac-tors in our experimental protocol. The gene sequences wereintroduced into the zygote at the site of the male pronucleus,allowing for the acceptance of these DNA sequences along withthe male chromosomal complement during the merger of thezygote pronuclei (14). Also, the rabbit 3-globin gene sequencein both the gene fragment and the intact Z-pCRI/Rchr(3G-1plasmid introduced into the mouse zygote pronucleus, and pre-viously shown to function and produce rabbit (3-globin mRNAin mouse L cells in culture (26), contained a 233-base-pair se-quence upstream from the cap site of the structural gene (19),which included the Hogness-Goldberg T-A-T-A (27) and theC-A-A-T sequences (28) believed to be important for translationof the genetic message in vivo (28). It also may be significantthat the gene introduced into the mice (i.e., rabbit (3-globin Z-pCRI/Rchrl3G-l or its Hha I fragment) shows extensive se-quence homology with the mouse /3-globin sequence (19).

Thefuill significance of the work reported here awaits furthercharacterization of the unique globin protein present in thetransformed mice and elaboration of the exact relationship ofthe introduced rabbit (-globin gene to the total mouse genomeas well as the fidelity with which it is expressed and transmittedto the progeny ofthe original recipient mice. However, it is notpremature to suggest that precise alteration of the genetic andphenotypic makeup of individual animals may now be possible.We thank Drs. R. Flavell and F. Grosveld for the kind gift of the Z-

pCRI/RchrXG-1 plasmid, Drs. Chen Jung Hsu and Jane E. Barker forvaluable discussion, and Ms. Catherine J. Knowles and A. W. Lun-gershausen for excellent technical assistance. We also wish to thank Dr.Chengbo Zhang, Visiting Professor from the Peoples Republic ofChina,and Dr. Bennett N. Cohen for invaluable assistance. This investigationwas supported, in part, by National Institutes of Health Grant HD10381 to P.C.H. The Jackson Laboratory is fully accredited by theAmerican Association for Accreditation of Laboratory Animal Care.

1. Wigler, M., Sweet, R., Sim, G. K., Wold, B., Pellicer, A., Lacy,E., Maniatis, T., Silverstein, S. & Axel, R. (1979) Cell 16,777-785.

2. Wigler, M., Pellicer, A., Silverstein, S. & Axel, R. (1978) Cell 14,725-731.

3. Wigler, M., Pellicer, A., Silverstein, S., Axel, R., Urlaub, G. &Chasin, L. (1979) Proc. Nat!. Acad. Sci. USA 76, 1373-1376.

4. Willecke, I., Klomfass, M., Mierau, R. & Dohmer, J. (1979) Mol.Gen. Genet. 170, 179-185.

5. Graf, L. H., Urlaub, G. & Chasin, L. (1979) Som. Cell Genet. 5,1031-1044.

6. Wigler, M., Perucho, M., Kurtz, D., Dana, S., Pellicer, A.,Axel, R. & Silverstein, S. (1980) Proc. Nat!. Acad. Sci. USA 77,3567-3570.

7. Anderson, W. F., Killos, L., Saunders-Haigh, L., Kretschmer,P. J. & Diacumakos, E. G. (1980) Proc. Nat!. Acad. Sci. USA 77,5399-5403.

8. Capecchi, M. (1980) Cell 22, 479-488.9. Jaenische, R. Mintz, B. (1974) Proc. Nat!. Acad. Sci. USA 71,

1250-1254.10. Gordon, J. W., Scangos, G. A., Plotkin, D. J., Barbosa, J. A. &

Ruddle, F. H. (1980) Proc. Nat!. Acad. Sci. USA 77, 7380-7384.11. Wagner, T. E., Sliwinski, J. E. & Shewmaker, D. B. (1978) Arch.

Andr. 1, 31-41.12. Sipski, M. L. & Wagner, T. E. (1977) Bio. Reprod. 16, 428-440.13. Sipski, M. L. & Wagner, T. E. (1977) Biopolymers 16, 573-582.14. Shapiro, B. M. & Eddy, E. M. (1980) Int. Rev Cyt. 66, 257-302.15. Kunkle, M., Longo, F. J. & Magum, B. E. (1978) J. Exp. Zoo!

203, 371-380.16. Kunkle, M., Magum, B. E. & Longo, F. J. (1978) J. Exp. Zool.

203, 381-390.17. Hoppe, P. C. & Pitts, S. (1973) Bio. Reprod. 8, 420-426.18. Hoppe, P. C. & Illmensee, K. (1977) Proc. Nat!. Acad. Sci. USA

74, 5657-5661.19. van den Berg, J., van Ooyen, A., Mantei, N., Schambock, A.,

Grosveld, G., Flavell, R. A. & Weissman, C. (1978) Nature (Lon-don) 276, 37-44.

20. Fanelli, A. R., Antonini, E. & Caputo, A. (1958) Biochim. Bio-phys. Acta 30, 608-615.

21. Whitney, J. B., Copland, G. T., Skow, L. C. & Russell, E. S.(1979) Proc. Natl Acad. Sci. USA 76, 867-871.

22. Jeffreys, A. J. & Flavell, R. A. (1977) Cell 21, 429-439.23. Southern, E. M. (1975) J. Mol. Biol. 98, 503-517.24. Summers, J. (1975) J. Virol. 15, 946-953.25. Weissman, C., Mantei, N., Boll, W., Weaver, R. F., Wilkie, N.,

Clements, B., Taniguchi, T., van Ooyen, A., van den Berg, J.,Fried, M. & Murray, K. (1979) From Gene to Protein: Informa-tion Transfer in Normal and Abnormal Cells (Academic) pp.99-132.

26. Mantei, N., Boll, W. & Weissmann, C. (1979) Nature (London)281, 40-46.

27. Goldberg, M. (1979) Dissertation, (Stanford Univ).28. Efstratiadis, A., Posakony, J. W., Maniatis, T., Lawn, R. M.

O'Connell, C., Spritz, R. A., DeRiel, J. K., Forget, B. G. Weiss-man, S. M., Slightom, J. L., Blechl, A. E., Smithies, O., Baralle,S. E., Shoulders, C. C. & Proudfoot, N. J. (1980) Cell 21,653-668.

6380 Genetics: Wagner et dD

ownl

oade

d by

gue

st o

n Ju

ly 2

8, 2

020