.=) 1992 Oxford University Press
Nucleic Acids Research, Vol. 20, No. 12 2971-2976
Targeted integration of neomycin into yeast artificial chromosomes (YACs) for transfection into mammalian cells J.H.Riley, J.E.N.Morten and R.Anand ICI Pharmaceuticals, Biotechnology Department, Alderley Park, Macclesfield, Cheshire SK10 4TG, UK Received April 10, 1992; Revised and Accepted May 15, 1992
ABSTRACT Vectors have been constructed for the introduction of the neomycin resistance gene (neo) into the left arm, right arm or human insert DNA of yeast artificial chromosomes (YACs) by homologous recombination. These vectors contain a yeast selectable marker Lys-2, i.e. the a-aminoadipidate reductase gene, and a mammalian selection marker, neo, which confers G418 resistance. The vectors can be used to modify YACs in the most commonly used yeast strain for YAC library construction, AB1 380. Specific targeting can be carried out by transfection of restriction endonuclease treated linear plasmids, with highly specific recombinogenic ends, into the YAC containing yeast cells. Analysis of targeted YACs confirmed that all three vectors can target correctly in yeast. Introduction of one of the targeted YACs into V79 (Chinese hamster fibroblast) cells showed complete and intact transfer of the YAC.
INTRODUCTION Until recently the study of gene expression by transfection of cloned DNA into eukaryotic cells was limited to the size of DNA fragments (-40kb) that could be cloned into cosmids. However, the introduction of YAC vectors (1) enabled the cloning of large genes covering up to 10OOkb, allowing associated upstream and downstream elements to be cloned as a single DNA fragment. This opened up the possibility of studying gene expression of large genes under the control of their own cis acting control elements. However, the most commonly used vector pYAC4 lacks a mammalian selectable marker, and therefore, one must be introduced if YACs are to be transfected into mammalian tissue culture cells. Introduction of YACs containing mammalian selectable genes into cell-lines has been achieved using complementation selection. This has been carried out with both HPRT(2) and GART(3) demonstrating transfer and functional expression of gene sequences from the YAC to the mammalian cells, for review see reference (4). There have also been several reports of modified vectors and strains allowing the use of different selection systems (5-8). The Ade-2 selective marker has been used to introduce neomycin resistance gene into YACs maintained in AB1380, but this resulted in multiple insertions at the target site (5) Targeting
of the mammalian selection system for thymidine kinase (TK) to a YAC using vectors containing Lys-2 as the yeast selectable marker and simultaneous knockout of Ura-3 limits its use to selection in TK negative lines (6). Further development of this system has lead to a single step process for targeting neo to the right hand arm of the YAC and a two step process for the modification of YACs at Line 1 (Ll) repetitive elements and the incorporation of neo (7). Finally, a different yeast strain YPH252 was used in which complementation of His-3 allowed targeting of a neomycin resistance gene to Alu repeats within the YACs (8). However, this could only be used for YACs which were either maintained in, or transferred to, the appropriate yeast
strain. Due to the limitations of some of the previous systems, we set out to develop vectors which would allow modification of any YAC in Saccharomyces cerivisiae AB1380, the most commonly used strain for YAC library construction, at one of three separate locations, using a single step process. This would also allow us to test the function of the mammalian selectable marker at these different sites. A yeast selectable marker was also incorporated on the targeting vectors to facilitate selection of cells integrating the incoming DNA. The yeast selection system we used was the Lys-2 gene obtained from the plasmid pDP-6 (9) and the mammalian selectable marker used was the neomycin resistance gene driven by a mouse phosphoglycerate-kinase promoter (10).
MATERIALS AND METHODS Restriction endonucleases and DNA modifying enzymes were obtained from Boehringer Mannheim and restriction digests were carried out according to the manufacturers instructions. All plasmid DNA was prepared by alkaline lysis followed by caesium chloride purification (11). Vector Construction Left hand targeting vector or Amp targeting vector. The Lys-2 containing plasmid pDP-6 (9) was used as the starting plasmid for this vector. A mouse phosphoglycerate kinase promoterdriven neomycin resistance gene (10) was excised from a bluescript vector (Stratagene) by restriction endonuclease treatment with SstI and EcoRI. This 2.0kb fragment was cloned
2972 Nucleic Acids Research, Vol. 20, No. 12 into SstI/ EcoRI restricted pDP6 to create the amp/left hand targeting vector pLNA-l (Lys, Neo, Amp-targeting), (Figure 1). For targeting into the left hand arm of the YAC, the vector was linearised with Aatll to create the recombinogenic ends. Following precipitation with 2.5M ammonium acetate and two volumes of 96% ethanol, and a 70% ethanol wash, the linearised DNA was resuspended in double distilled water. Transformation was carried out as described later using 2jLg of the DNA per transfection.
Right hand targeting vector or Tet targeting vector. The plasmid pLNA-l was digested with SmaI and dephosphorylated with an excess of calf intestinal alkaline phosphatase 2U/4g of DNA since this was done in the restriction digest buffer. The targeting fragment containing part of the right hand arm, which possessed an incomplete tetracycline resistance gene(l.9kb), was excised from pYAC4 by digestion with SmaI/ ScaL. This was then ligated to SmaI restricted pLNA-l to create the vector pLNT-l (Lys, Neo, Tet-targeting), (Figure2). The vector was targeted into the right hand of the YAC by linearising with EcoNI, which created the recombinogenic ends. The vector was prepared for transfection as described above for pLNA-l. Human Alu targeting vector. The pLNA-l vector was digested with EcoRI, the site was filled-in with T4 DNA polymerase, following phenol-chloroform extraction and precipitation with ethanol, dephosphorylated with calf intestinal alkaline phosphatase according to Boehringer Mannheim instructions. The 0.3kb Blurl3 Alu repeat was removed from pBP47 (8) by digestion with BamHI. After treatment with T4-DNA Polymerase to create blunt ends it was ligated into the vector to create pLNB-l(Lys, A
-r
;'
.-'*EF
tANtU AFB
Neo, Blur-targeting), (Figure3). Creation of the recombinogenic ends for pLNB- I was achieved by EcoRI digestion of the DNA. Yeast Culture and Transformation The two YACs used for these targeting studies were: CFTR YAC 37AB12 (310kb), MATa, ade2-1, canl-100, lys2-1, his-5. (12). DR4,B YAC 410A3 (220kb) MATa, ade2-1, can1-100, lys2-1, his-5 obtained from C.E.P.H, France. The YACs were grown in synthetic dextrose medium (0.7% yeast nitrogen base without amino acids, 1 % D-glucose, 5.5mg adenine hemisulphate and 5.5 mg tyrosine per lOOml) containing 7ml of 20% cas amino acids per lOOml, which lacks both tryptophan and uracil as described (13). Transformation with 21tg of linearised DNA was carried out using yeast spheroplasts prepared with lyticase, and using solutions as described by Burgers and Percival(14). Transformants were selected on synthetic dextrose plates lacking tryptophan, uracil and lysine in the 3% top agar and were visible after 3 to 5 days.
Electrophoresis and Hybridisation Preparation of high molecular weight cell-line DNA was carried out according to Anand and Southden (15). High molecular weight DNA from yeast cells for PFGE analysis and normal southern analysis was prepared in plugs using lithium dodecyl sulphate (15). PFGE analyses were carried out on a Waltzer PFGE apparatus (Hula gel, Hoeffer). The gels were blotted onto Zetaprobe GT membrane (Biorad) and hybridisation was according to the manufacturers instructions. Probes used for the analyses were pBR322 LHE (Ampicillin gene), pBR322 RHE (Tetracycline gene) which correspond to the left and right arms
JM. lF; !
/~2~\
_
_
~~~~~~~~..-
lioroiogous Reglr Aatl.
r-
IOLNA f.'3 k
vs
ruki
B.
.dI-------*
--Hr-.'-
-. K
z .dK *e.
FIgure 1. Modification of YAC 410A3 using pLNA-1. IA. The YAC modification vector pLNA-1 was digested with Aatll to create recombinogenic ends. The region of homology between the vector and the YAC left hand arm is shaded. lB. Hybridisation of the blots of EcoRI digested YACs with neo, lys-2 and pBR322 LHE demonstrated two types of recombination event. Lane 1 in each case shows single insertion of the vector at the AatII site. Lane 2 in each case shows multiple insertion of the vector at the Aatll site on the left hand arm of the YAC with the extra band at 9.5kb representing the size of the vector, (see text for details). IC. The vector pLNA-1 inserts by recombination at the Aatll site to give the targeted YAC as shown. The fragments obtained by hybridising the gel with the various probes are shown underneath the modified YAC diagram.
Nucleic Acids Research, Vol. 20, No. 12 2973 of the YAC respectively (1); a 2.0kb EcoRI/ Sstl neomycin probe from pLNA-1 and a 4.5kb lys-2 probe removed by Sall digests of pDP6. Probes for the CFTR exonsl9, 23 and 13, were as described in (12) being prepared by PCR amplification, followed by purification and labelling in low gelling temperature agarose. D9, a probe which mapped to the extreme left hand end of the YAC (37AB12) was also synthesised by PCR (16). Labelling of probes was by random priming (17).
RESULTS Targeting of pLNA-1 to YACs Following transformation of YACs 410A3 and 37AB12 with pLNA-1 (FigurelA), 20 to 100 colonies per jAg of linearised DNA were observed on selective medium lacking lysine. Initial analysis of the recombinants by PFGE and Southern hybridisation with a neo probe showed that for 410A3,10/20 of the recombinants possessed the neo gene (results not shown), whilst for 37AB12, 8/18 contained neo. The remainder of the recombinants which did not contain neo are thought to represent a correction of the yeast host (AB1380) mutant Lys-2 gene, from the functional Lys-2 gene on the introduced plasmid. These colonies were not due to reversions as control plates of cells only, gave no colonies suggesting a very low reversion frequency. Southern analysis of the PFGE gels of neo positive recombinants demonstrated that neo was on the artificial chromosome (data not shown). Digestion of these neo containing clones with EcoRI and hybridisation of the blots with neo showed that in half of the samples there was a band hybridising at 13kb representing correct insertion into the left hand arm. In the other half there were two bands visible (see FigureIB), the upper band of 13kb represents the integration event we expected via the recombinogenic ends created at the AatlI site (see Figure IC) and the lower band of 9.5kb representing the size of the vector. This represents a multiple integration event at the Aatll site. Repeat hybridisation of the blot with lys-2 showed two bands in the single insertion event, one band at 13kb, i.e. the same as with the neo hybridisation, and an additional band at 15kb representing the yeast mutant Lys-2 gene present in the host Saccharomyces cerivisiae AB1380. In the multiple insertion events, the same bands are visible as
Cell Culture and Fusion YACs possessing the correct genotype (ura+, trp+ and lys+) were grown overnight at 30°C in media lacking typtophan, uracil and lysine. The cells were then diluted in 100 ml, to an absorbance of lOD/ml at 600 nm, grown for two doublings at 30°C in YPD (2% glucose, 2% bactopeptone and 1% yeast extract) and harvested. Cells were washed in water, recovered by centrifugation at i5OOg at room temperature, re-washed in IM sorbitol, recovered and spheroplasted in 50 ml, using 2000U lyticase. Cells were re-washed twice in IM sorbitol and finally resuspended in IM sorbitol at 2 x 107 cells/ml. The recipient cells for the spheroplast fusion were Chinese hamster fibroblasts, V79A2341 (HPRTnegative and obtained from J.Boyle, Patterson Institute For Cancer Research, Manchester). Exponentially growing cells were trypsinised and washed three times in serum free DMEM (Gibco/ BRL). The procedure for the fusion was essentially as described in (2), except that we used 2 x 108 spheroplasts per 2 x 106 recipients and the times in PEG were 1, 2 and 3 minutes before adding 5mls serum free DMEM. G418 selection was applied 48 hours after cell fusion. YAC RIGHT HAND ARM. HUMAN DNA
28
EcoRl EcoNI II-
23.1
a
-WA.
MA&
Homologous Region EcoNI
9.4
NEO
6.6
qSD
Notl 1
/\a
2
LYS-2
pLNT-1 .LYS-2
3
2
2 3 3 NEO pBR322 RHE
11.-5Kb
1.
EcoRl
1-
AMP
2C EcoRI HUMAN DNA
EcoRI
LYS-2
1~
AMP
NEO
eN= ~~~~~~~~~~~~ LYS2
9.5Kb
6.0Kb
pBR322 RHE
NEO
Figure 2. Modification of YAC 410A3 using pLNT-l. 2A. The YAC modification vector pLNT-1 was digested with EcoNI to create recombinogenic ends. The region of homology between the vector and the YAC right hand arm is shaded. 2B. Hybridisation of the blots of EcoRl digested YACs with neo, lys-2 and pBR322 RHE demonstrates two different types of recombination events. Lane I is only shown hybridised with lys-2 and is an EcoRI digest of DNA from the host strain AB1380 showing the non functional lys-2 gene at 15kb. Lane 2 in each case shows single insertion of the vector at the EcoNI site. Lane 3 in each case shows multiple insertion of the vector at the EcoNI site on the right hand arm of the YAC with the extra band at 11.5kb representing the size of the vector (see text for details). 2C. The vector pLNT-1 inserts by recombination at the EcoNI site to give the targeted YAC as shown. The fragments obtained by hybridising the gel with the various probes is shown underneath the modified YAC diagram.
2974 Nucleic Acids Research, Vol. 20, No. 12
*d
in the single insertion event an in addition-the extra band at 9.5kb representing the size of the vector. On hybridisation of the blot with pBR322 LHE there are two bands visible with the single integrants, the band at 13kb was again visible as expected, and a smaller band at 3.8kb which is the di&tance from the EcoRI site in the pLNA-1 to the end of the YAC telomere see (Figure IC). In the multiple integration events there is the addi band at 9.5kb, again representing the size of the vector.
insertion event there are three bands visible at 6kb, 9.5kb (as in the single insertion event) and an extra band at 11.5kb representing the size of the vector. Hybridising the same blot with lys-2 shows the endogenous Lys-2 gene at 15kb, the fragment at 9.5kb represents the Lys-2 gene integrated correctly into the EcoNI site of the YAC arm. The extra band at 11.5kb in the multiple integration event represents the size of the vector following multiple insertion at the EcoNI site.
Targetig of pLNT-1 to YACs Following trasformation of YACs 410A3 and 37AB12 with linearised pLNT-1 (Figure 2A), again approximately 20-100 colonies per pg were observed on selective medium lacking lysine. Initial analysis of the recombinants by PFGE and hybridisation of the blot with neo showed for 410A3, 9/18 of the recombinants possessed neo and for 37AB12, 11/20 possessed neo (results not shown). Again the remainder not containing neo s resting from repair of the mutant are dtught to be Lys-2 gene in AB1380. TMe neo positive recombinants were analyzed fiuther by digeston with EcoRI (see Figure 2B). On hybridising the blot with neo as before either one or two bands were visible. Ihe lower band at 6kb repesents correct in at the EcoNI site of the right hand arm, whilst the upper band at 11.5kb represents the size of the vector following a multiple integration event. Six recombinants only psessed the lower band at 6kb representing correct integration at the EcoNI site in the right hand arm. On hybridising the same blot with pBR322 RHE, in the single integraton event there are two bands visible. The band at 6kb represents the same band as in the neo hybridisation, the band at 9.5kb represents the rest of the vector and some right hand arm sequences as expected (see Figure 2C). In the multiple
Targeting of pLNB-1 to YACs Yeast cells containing the 220kb YAC 410A3 were transformed with linearised pLNB-1 (Figure 3A). Initial analysis of 10 colonies growing on media lacking lysine, by hybridisation of a PFGE blot with neo showed that four of the clones had targeted to the correct size of the YAC whilst one had neo targeted to the Lys-2 gene present in the yeast and one had neo targeted at an increased size (data not shown). The odrer four of these clones had no neo present on hybridisation and presumably represented repair of the endogenous mutant Lys-2 gene (data not shown). Further PFGE analysis was carried out following digestion of the recombinants with Not. The pLNB-1 vector contains a Notd site which on integration into the YAC should introduce the site into the human DNA of the YAC. On hybridisation with neo three of the recombinants showed integration into the left hand arm presumably by homology of the ampicillin resistance gene present in the vector and in the left hand arm of the YAC (see Figure 3B track 2). One recombinant had multiple insertions at different locations in the human DNA (data not shown), one clone had a single integration event but at the lys-2 gene (data not shown) and one clone had a single integration event showing a single band hybridising at 190kb (Figure 3B track 1). Further
3A
-
ii
Cl 0
-f.f)
A'
'S
::
CO'
AM-
:?-
.s HI
NB-
;BK
Lett
*.* .;z * 'y'a-_a
a
S
*
M c.t NE: Arir: E
- OK-t
F%gur
\
_ . o,
3. Modification of YAC 410A3 using PLNB-1. 3A. The YAC modification vector pLNB-1 was digested with EcoRI to creat recombinogenic ends. The region of homology between fth vector and the YAC is shaded. 3B. Not I digested YAC DNA was blotted and hybridsed with various probes. Lane 1 yAC targete by insertion into the centre of the humian DNA, Lane 2 YAC tageted by insertion into the left hand arm of the yAC spuriously. Lane 3, undigested, tarete YAC showing the non functiona Lys-2 gene at limiting mobility and the targeted YAC at 230kb. 3C. The vector pLNB-1 inserts by recomination at the EcoRI site to give the targeted YAC as shown. The fragments obtained by hybridising the gel with the various probes are shown underneath the modified YAC diagram.
Nucleic Acids Research, Vol. 20, No. 12 2975 analysis of this clone by hybridisation of the blot with lys-2 showed two bands. The band at approximately 450kb represents the non-functional Lys-2 gene present in the yeast following NotI digestion and the band hybridising at 40kb represented the fragment from the other side of the NotI site (Figure 3C). On hybridisation with pBR322 LHE both the bands previously seen hybridised i.e. at 190kb and 40kb. Following repeat hybridisation with of the blot with pBR322 RHE a band of 40kb was visible (Figure 3B). From the data obtained it was possible to show the orientation of the insertion (see Figure 3C).
Fusion of 37AB12-pLNA-1 with V79 cells Approximately 1 in 104 to 105 G418 resistant V79 clones were obtained following fusion with 37AB12 pLNA-1. Genomic DNA was restricted with EcoRI, blotted and hybridised with pBR322, lys-2, CFTR exons 19, 23 and D9. Hybridisation with pBR322 showed multiple bands indicating the transfer of YAC arm sequences (Figure 4A). The band at 13kb was indicative of the transfer of the left hand arm sequences as shown previously (Figure IB) in the targeted YAC. To check if there was transfer of yeast sequences the blot was rehybridised with lys-2. As previously mentioned, the AB1380 strain possesses a non functional Lys-2 gene shown by a band at 15kb (Figure 4B). In four out of the five cell lines analyzed this yeast band was shown to be present suggesting that at least part, if not all of the yeast genome was transferred to the cell-lines. The other band hybridising with lys-2 at 13kb shows transfer of the left hand arm sequences from the YAC. On hybridising the blot with D9, CFTR exon 19 and exon 23 there were three bands hybridising at 4.0kb (D9), 2.2kb (CFTR 23) and at 2.4kb (CFTR 19) as shown in (Figure 4C). This confirms the transfer of CFTR sequences from the extreme left hand end of the YAC (D9) and the right hand end of the YAC(exon 19, 23). To confirm the presence of the central portion of the YAC SacII digests were carried out and the fragments resolved by PFGE. A band hybridising to a CFTR exon 13 probe of 2 10kb showed complete transfer of this region (Figure 4D).
4A
4B
DISCUSSION We have demonstrated that using the vectors described in this paper, it is possible to modify YACs in Saccharomyces cerivisiae (AB1380) to introduce a phosphoglycerate kinase driven neomycin resistance gene and the yeast selectable marker for aaminoadipidate reductase (Lys-2). Targeting of the linearised vectors into the YACs by homologous recombination has been shown to be highly specific for the left and right hand arm of the YAC allowing the introduction of one or more copies of the neo gene. However, the alu targeting vector has been found to be less specific, producing artefactual integrations into the left hand arm of the YAC and into the mutated Lys-2 gene of Saccharomyces cerivisiae. The alu targeting vector only targets tol5Obp on either side of the EcoRI site which on restriction digestion creates the recombinogenic ends. Since the vector is 11.5kb in length, there is scope for incorrect targeting. Furthermore the alu repeats within the YAC may not possess perfect sequence homology or correct orientation for targeting, resulting in few correct targeting events. Nevertheless we did obtain one out of ten clones which had integrated correctly at an alu repeat. The use of LI line repeat sequences has been found to work better than the alu repeat (7). Both the left hand and right hand targeting vectors integrate by insertion whereas the alu targeting vector can integrate by insertion or by replacement. From our data, it is apparent that the recombination event we observed was by insertion since there was no reduction in size of the YAC following targeting. Replacement can cause deletions in the YAC insert DNA by recombination between two repetitive alu elements which lie in close proximity, as has been observed in other clones we have analyzed (J.Riley, unpublished). A significant background of neo negative, lys positive recombinants (approximately 50%) was observed in all cases of modification. This is likely to be due to recombination repair of the mutant Lys-2 gene in the AB1380 host. This is easily screened for by hybridising filter lifts of lys positive clones with a neo probe to find the lys positive, neo positive colonies.
4C
4D
23.1 -
I
.
9.4
k- 23.1
--- 6.6
.4
---4.4
6.6 .4.4
l 1
2
pBR322
2 1
LYS-2
2
CFTR 23, 19 and D9
CFTR 13
Figure 4. Fusion of the modified CFTR YAC (37AB12-pLNA-1) with V79 cells. Two independent clones obtained after spheroplast fusion of 37AB12-pLNA-l to V79 cells (Chinese hamster fibroblasts). 4A. Lanes 1 and 2 represent EcoRI restricted DNA from the two cloned cell lines following blotting and hybridisation with pBR322. The band at 13kb is indicative of transfer of the left hand arm. 4B. Hybridisation of the two clones with lys-2 showed transfer of the 13kb pBR322 fragment as before and transfer of the 15kb non functional (AB1380) lys-2 gene. 4C. Hybridisation of the two clones with CFTR exons 23,19 and D9 showed the presence of all three fragments. 4D. Finally following SstII digestion of the two clones, fractionation by PFGE and hybridisation with CFTR exonl3 showed a band hybridising at 220kb, indicative of the transfer of the majority of the CFTR gene as a single DNA fragment.
2976 Nucleic Acids Research, Vol. 20, No. 12 The YACs used for these studies were 37AB12 which contained the whole of the CFTR gene and 410A3 containing DR4,B haplotype and were obtained from two independent libraries constructed using the vector pYAC-4 and were in the yeast host Saccharomyces cerivisiae AB1380. The efficient recombination systems in yeast can also be used to generate mutations by co-transfection of a modified mutation-containing fragment with a selection cassette. For CFTR this may include the deletion of amino acid 508 (Phenylalanine) (18), which is the most commonly found mutation in cystic fibrosis patients. Neo positives may be tested for the correct integration of the mutation by the use of allele specific PCR or Amplification Refractory Mutation System, ARMS technology (19). The introduction of YACs into cell-lines by fusion is already proving to be a useful tool for the study of gene function (2,3). It may in the foreseeable future prove useful in the construction of transgenic models of human disease via the Embryonic Stem cell route. The modified YAC 37AB12-pLNA-l described here was initially introduced into V79 Chinese hamster fibroblast cells using cell fusion. The presence of G418 resistant colonies showed that the PGK-neo functioned well in the cell lines. The results demonstrated that complete transfer of the YAC had occurred and that both the YAC arms were transferred, along with the yeast DNA. To test if the incoming yeast DNA integrates randomly at multiple sites within the yeast genome, experiments using fluorescent in situ hybridisation were carried out (J.Morten unpublished). These studies clearly indicated that although the site of integration was apparently random, most, if not all of the incoming yeast DNA goes into the Chinese hamster fibroblast chromosomes in one block. This suggests that the incoming yeast DNA is unlikely to disrupt a large number of host genes. These vectors will prove a useful tool in modifying YACs to incorporate a mammalian selectable marker, for studies on gene expression in cell lines, where a large gene is under the control of its own cis acting factors.
ACKNOWLEDGEMENTS We would like to thank Dr P.Philippsen for the gift of pDP-6 and Denis Le Paslier for the gift of YAC 410A3. We would also like to thank the Biographics department at ICI Pharmaceuticals for the photographs appearing in this manuscript.
REFERENCES 1. Burke, D.T. , Carle, G.F. and Olson, M.V. (1987). Science., 236, 806-812. 2. Huxley,C., Hagino, Y., Schlessinger,D. and Olson,M.V. (1991). Genomics, 9, 742-750. 3. Gnirke, A., Barnes, T.S., Patterson, D., Schild, D., Featherstone,T. and Olson, M.V. (1991). EMBO J., 10, 1629-1634. 4. Huxley, C. and Gnirke, A. (1991). Bioessays, 13, 545-549. 5. Pachnis, V., Pevny, L., Rothstein, R. and Constantini, F. (1990). Proc. Natl. Acad. Sci. USA., 87, 5109-5113. 6. Eliceiri, B., Labella, T., Hagino, Y., Srivastava, A., Schlessinger,D., Pilia,G., Palmieri, G. and D'Urso, M. (1991). Proc. Natl. Acad. Sci. USA.,
88, 2179-2183.
7. Srivastava A, K. and Schlessinger, D. (1991). Gene, 103, 53-59. 8. Pavan, W.J., Hieter, P. and Reeves, R.H. (1990). Mol.Cell.Biol., 10,
4163-4169. 9. Fleig, U.N., Pridmore, R.D. and Philippsen, P. (1986). Gene, 46, 237-245. 10. Adra, C.N., Boer, P.H. and McBurnay, M.W. (1987). Gene, 60, 65-74. 11. Molecular Cloning A Laboratory Manual 2nd Edition. Eds Sambrook,J., Fritsch,E.F. and Maniatis,T.(1989). Cold Spring Harbour Press. ppl.38, 1.39 and 1.42. 12. Anand, R., Ogilvie, D.J., Butler, R., Riley, J.H., Finniear, R.S., Powell,S., Smith,J.C and Markham, A.F. (1991). Genomics, 9, 124-130.
13. Anand, R., Riley, J.H., Butler, R., Smith, J.C. and Markham, A.F. (1990).
Nucl. Acid, Res.,18, 1951-1956. 14. Burgers, P. M. J. and Percival, K. J. (1987). Anal. Biochem., 163, 391-397. 15. Gel Electrophoresis of Nucleic Acids; A Practical Approach. Eds Rickwood, D. and Hames, B, D. IRL Press. pp 101-123. Anand, R. and Southern, E, M. (1989). 16. Estivill, X., McLean,C., Nunes, V., Casals, T., Gallano,P., Scambler, P. and Williamson, R. (1989). Am.J. Hum. Genet., 44, 704-710. 17. Feinberg, A,P. and Vogelstein, B. (1983). Anal. Biochem., 132, 6-13. 18. Riordan, J. R., Rommens, J. H., Kerem, B.S., Alon, N., Rozmahel, R.,
Grzelczak, Z., Zielenski, J., Lok, S., Plavsic, N., Chou, J.L., Drumm, M.L., lannuzzi, M.C., Collins, F.S., and Tsui, L.C. (1989). Science., 24s,
1066-1073. 19. Newton, C.R., Graham, A., Heptinstall, L.E., Powell, S.J., Summers, C., Kalsheker, N., Smith, J.C., and Markham, A,F. (1989). Nucl. Acid, Res., 17, 2593-2516.
Nucleic Acids Research, Vol. 20, No. 12 2971-2976
Targeted integration of neomycin into yeast artificial chromosomes (YACs) for transfection into mammalian cells J.H.Riley, J.E.N.Morten and R.Anand ICI Pharmaceuticals, Biotechnology Department, Alderley Park, Macclesfield, Cheshire SK10 4TG, UK Received April 10, 1992; Revised and Accepted May 15, 1992
ABSTRACT Vectors have been constructed for the introduction of the neomycin resistance gene (neo) into the left arm, right arm or human insert DNA of yeast artificial chromosomes (YACs) by homologous recombination. These vectors contain a yeast selectable marker Lys-2, i.e. the a-aminoadipidate reductase gene, and a mammalian selection marker, neo, which confers G418 resistance. The vectors can be used to modify YACs in the most commonly used yeast strain for YAC library construction, AB1 380. Specific targeting can be carried out by transfection of restriction endonuclease treated linear plasmids, with highly specific recombinogenic ends, into the YAC containing yeast cells. Analysis of targeted YACs confirmed that all three vectors can target correctly in yeast. Introduction of one of the targeted YACs into V79 (Chinese hamster fibroblast) cells showed complete and intact transfer of the YAC.
INTRODUCTION Until recently the study of gene expression by transfection of cloned DNA into eukaryotic cells was limited to the size of DNA fragments (-40kb) that could be cloned into cosmids. However, the introduction of YAC vectors (1) enabled the cloning of large genes covering up to 10OOkb, allowing associated upstream and downstream elements to be cloned as a single DNA fragment. This opened up the possibility of studying gene expression of large genes under the control of their own cis acting control elements. However, the most commonly used vector pYAC4 lacks a mammalian selectable marker, and therefore, one must be introduced if YACs are to be transfected into mammalian tissue culture cells. Introduction of YACs containing mammalian selectable genes into cell-lines has been achieved using complementation selection. This has been carried out with both HPRT(2) and GART(3) demonstrating transfer and functional expression of gene sequences from the YAC to the mammalian cells, for review see reference (4). There have also been several reports of modified vectors and strains allowing the use of different selection systems (5-8). The Ade-2 selective marker has been used to introduce neomycin resistance gene into YACs maintained in AB1380, but this resulted in multiple insertions at the target site (5) Targeting
of the mammalian selection system for thymidine kinase (TK) to a YAC using vectors containing Lys-2 as the yeast selectable marker and simultaneous knockout of Ura-3 limits its use to selection in TK negative lines (6). Further development of this system has lead to a single step process for targeting neo to the right hand arm of the YAC and a two step process for the modification of YACs at Line 1 (Ll) repetitive elements and the incorporation of neo (7). Finally, a different yeast strain YPH252 was used in which complementation of His-3 allowed targeting of a neomycin resistance gene to Alu repeats within the YACs (8). However, this could only be used for YACs which were either maintained in, or transferred to, the appropriate yeast
strain. Due to the limitations of some of the previous systems, we set out to develop vectors which would allow modification of any YAC in Saccharomyces cerivisiae AB1380, the most commonly used strain for YAC library construction, at one of three separate locations, using a single step process. This would also allow us to test the function of the mammalian selectable marker at these different sites. A yeast selectable marker was also incorporated on the targeting vectors to facilitate selection of cells integrating the incoming DNA. The yeast selection system we used was the Lys-2 gene obtained from the plasmid pDP-6 (9) and the mammalian selectable marker used was the neomycin resistance gene driven by a mouse phosphoglycerate-kinase promoter (10).
MATERIALS AND METHODS Restriction endonucleases and DNA modifying enzymes were obtained from Boehringer Mannheim and restriction digests were carried out according to the manufacturers instructions. All plasmid DNA was prepared by alkaline lysis followed by caesium chloride purification (11). Vector Construction Left hand targeting vector or Amp targeting vector. The Lys-2 containing plasmid pDP-6 (9) was used as the starting plasmid for this vector. A mouse phosphoglycerate kinase promoterdriven neomycin resistance gene (10) was excised from a bluescript vector (Stratagene) by restriction endonuclease treatment with SstI and EcoRI. This 2.0kb fragment was cloned
2972 Nucleic Acids Research, Vol. 20, No. 12 into SstI/ EcoRI restricted pDP6 to create the amp/left hand targeting vector pLNA-l (Lys, Neo, Amp-targeting), (Figure 1). For targeting into the left hand arm of the YAC, the vector was linearised with Aatll to create the recombinogenic ends. Following precipitation with 2.5M ammonium acetate and two volumes of 96% ethanol, and a 70% ethanol wash, the linearised DNA was resuspended in double distilled water. Transformation was carried out as described later using 2jLg of the DNA per transfection.
Right hand targeting vector or Tet targeting vector. The plasmid pLNA-l was digested with SmaI and dephosphorylated with an excess of calf intestinal alkaline phosphatase 2U/4g of DNA since this was done in the restriction digest buffer. The targeting fragment containing part of the right hand arm, which possessed an incomplete tetracycline resistance gene(l.9kb), was excised from pYAC4 by digestion with SmaI/ ScaL. This was then ligated to SmaI restricted pLNA-l to create the vector pLNT-l (Lys, Neo, Tet-targeting), (Figure2). The vector was targeted into the right hand of the YAC by linearising with EcoNI, which created the recombinogenic ends. The vector was prepared for transfection as described above for pLNA-l. Human Alu targeting vector. The pLNA-l vector was digested with EcoRI, the site was filled-in with T4 DNA polymerase, following phenol-chloroform extraction and precipitation with ethanol, dephosphorylated with calf intestinal alkaline phosphatase according to Boehringer Mannheim instructions. The 0.3kb Blurl3 Alu repeat was removed from pBP47 (8) by digestion with BamHI. After treatment with T4-DNA Polymerase to create blunt ends it was ligated into the vector to create pLNB-l(Lys, A
-r
;'
.-'*EF
tANtU AFB
Neo, Blur-targeting), (Figure3). Creation of the recombinogenic ends for pLNB- I was achieved by EcoRI digestion of the DNA. Yeast Culture and Transformation The two YACs used for these targeting studies were: CFTR YAC 37AB12 (310kb), MATa, ade2-1, canl-100, lys2-1, his-5. (12). DR4,B YAC 410A3 (220kb) MATa, ade2-1, can1-100, lys2-1, his-5 obtained from C.E.P.H, France. The YACs were grown in synthetic dextrose medium (0.7% yeast nitrogen base without amino acids, 1 % D-glucose, 5.5mg adenine hemisulphate and 5.5 mg tyrosine per lOOml) containing 7ml of 20% cas amino acids per lOOml, which lacks both tryptophan and uracil as described (13). Transformation with 21tg of linearised DNA was carried out using yeast spheroplasts prepared with lyticase, and using solutions as described by Burgers and Percival(14). Transformants were selected on synthetic dextrose plates lacking tryptophan, uracil and lysine in the 3% top agar and were visible after 3 to 5 days.
Electrophoresis and Hybridisation Preparation of high molecular weight cell-line DNA was carried out according to Anand and Southden (15). High molecular weight DNA from yeast cells for PFGE analysis and normal southern analysis was prepared in plugs using lithium dodecyl sulphate (15). PFGE analyses were carried out on a Waltzer PFGE apparatus (Hula gel, Hoeffer). The gels were blotted onto Zetaprobe GT membrane (Biorad) and hybridisation was according to the manufacturers instructions. Probes used for the analyses were pBR322 LHE (Ampicillin gene), pBR322 RHE (Tetracycline gene) which correspond to the left and right arms
JM. lF; !
/~2~\
_
_
~~~~~~~~..-
lioroiogous Reglr Aatl.
r-
IOLNA f.'3 k
vs
ruki
B.
.dI-------*
--Hr-.'-
-. K
z .dK *e.
FIgure 1. Modification of YAC 410A3 using pLNA-1. IA. The YAC modification vector pLNA-1 was digested with Aatll to create recombinogenic ends. The region of homology between the vector and the YAC left hand arm is shaded. lB. Hybridisation of the blots of EcoRI digested YACs with neo, lys-2 and pBR322 LHE demonstrated two types of recombination event. Lane 1 in each case shows single insertion of the vector at the AatII site. Lane 2 in each case shows multiple insertion of the vector at the Aatll site on the left hand arm of the YAC with the extra band at 9.5kb representing the size of the vector, (see text for details). IC. The vector pLNA-1 inserts by recombination at the Aatll site to give the targeted YAC as shown. The fragments obtained by hybridising the gel with the various probes are shown underneath the modified YAC diagram.
Nucleic Acids Research, Vol. 20, No. 12 2973 of the YAC respectively (1); a 2.0kb EcoRI/ Sstl neomycin probe from pLNA-1 and a 4.5kb lys-2 probe removed by Sall digests of pDP6. Probes for the CFTR exonsl9, 23 and 13, were as described in (12) being prepared by PCR amplification, followed by purification and labelling in low gelling temperature agarose. D9, a probe which mapped to the extreme left hand end of the YAC (37AB12) was also synthesised by PCR (16). Labelling of probes was by random priming (17).
RESULTS Targeting of pLNA-1 to YACs Following transformation of YACs 410A3 and 37AB12 with pLNA-1 (FigurelA), 20 to 100 colonies per jAg of linearised DNA were observed on selective medium lacking lysine. Initial analysis of the recombinants by PFGE and Southern hybridisation with a neo probe showed that for 410A3,10/20 of the recombinants possessed the neo gene (results not shown), whilst for 37AB12, 8/18 contained neo. The remainder of the recombinants which did not contain neo are thought to represent a correction of the yeast host (AB1380) mutant Lys-2 gene, from the functional Lys-2 gene on the introduced plasmid. These colonies were not due to reversions as control plates of cells only, gave no colonies suggesting a very low reversion frequency. Southern analysis of the PFGE gels of neo positive recombinants demonstrated that neo was on the artificial chromosome (data not shown). Digestion of these neo containing clones with EcoRI and hybridisation of the blots with neo showed that in half of the samples there was a band hybridising at 13kb representing correct insertion into the left hand arm. In the other half there were two bands visible (see FigureIB), the upper band of 13kb represents the integration event we expected via the recombinogenic ends created at the AatlI site (see Figure IC) and the lower band of 9.5kb representing the size of the vector. This represents a multiple integration event at the Aatll site. Repeat hybridisation of the blot with lys-2 showed two bands in the single insertion event, one band at 13kb, i.e. the same as with the neo hybridisation, and an additional band at 15kb representing the yeast mutant Lys-2 gene present in the host Saccharomyces cerivisiae AB1380. In the multiple insertion events, the same bands are visible as
Cell Culture and Fusion YACs possessing the correct genotype (ura+, trp+ and lys+) were grown overnight at 30°C in media lacking typtophan, uracil and lysine. The cells were then diluted in 100 ml, to an absorbance of lOD/ml at 600 nm, grown for two doublings at 30°C in YPD (2% glucose, 2% bactopeptone and 1% yeast extract) and harvested. Cells were washed in water, recovered by centrifugation at i5OOg at room temperature, re-washed in IM sorbitol, recovered and spheroplasted in 50 ml, using 2000U lyticase. Cells were re-washed twice in IM sorbitol and finally resuspended in IM sorbitol at 2 x 107 cells/ml. The recipient cells for the spheroplast fusion were Chinese hamster fibroblasts, V79A2341 (HPRTnegative and obtained from J.Boyle, Patterson Institute For Cancer Research, Manchester). Exponentially growing cells were trypsinised and washed three times in serum free DMEM (Gibco/ BRL). The procedure for the fusion was essentially as described in (2), except that we used 2 x 108 spheroplasts per 2 x 106 recipients and the times in PEG were 1, 2 and 3 minutes before adding 5mls serum free DMEM. G418 selection was applied 48 hours after cell fusion. YAC RIGHT HAND ARM. HUMAN DNA
28
EcoRl EcoNI II-
23.1
a
-WA.
MA&
Homologous Region EcoNI
9.4
NEO
6.6
qSD
Notl 1
/\a
2
LYS-2
pLNT-1 .LYS-2
3
2
2 3 3 NEO pBR322 RHE
11.-5Kb
1.
EcoRl
1-
AMP
2C EcoRI HUMAN DNA
EcoRI
LYS-2
1~
AMP
NEO
eN= ~~~~~~~~~~~~ LYS2
9.5Kb
6.0Kb
pBR322 RHE
NEO
Figure 2. Modification of YAC 410A3 using pLNT-l. 2A. The YAC modification vector pLNT-1 was digested with EcoNI to create recombinogenic ends. The region of homology between the vector and the YAC right hand arm is shaded. 2B. Hybridisation of the blots of EcoRl digested YACs with neo, lys-2 and pBR322 RHE demonstrates two different types of recombination events. Lane I is only shown hybridised with lys-2 and is an EcoRI digest of DNA from the host strain AB1380 showing the non functional lys-2 gene at 15kb. Lane 2 in each case shows single insertion of the vector at the EcoNI site. Lane 3 in each case shows multiple insertion of the vector at the EcoNI site on the right hand arm of the YAC with the extra band at 11.5kb representing the size of the vector (see text for details). 2C. The vector pLNT-1 inserts by recombination at the EcoNI site to give the targeted YAC as shown. The fragments obtained by hybridising the gel with the various probes is shown underneath the modified YAC diagram.
2974 Nucleic Acids Research, Vol. 20, No. 12
*d
in the single insertion event an in addition-the extra band at 9.5kb representing the size of the vector. On hybridisation of the blot with pBR322 LHE there are two bands visible with the single integrants, the band at 13kb was again visible as expected, and a smaller band at 3.8kb which is the di&tance from the EcoRI site in the pLNA-1 to the end of the YAC telomere see (Figure IC). In the multiple integration events there is the addi band at 9.5kb, again representing the size of the vector.
insertion event there are three bands visible at 6kb, 9.5kb (as in the single insertion event) and an extra band at 11.5kb representing the size of the vector. Hybridising the same blot with lys-2 shows the endogenous Lys-2 gene at 15kb, the fragment at 9.5kb represents the Lys-2 gene integrated correctly into the EcoNI site of the YAC arm. The extra band at 11.5kb in the multiple integration event represents the size of the vector following multiple insertion at the EcoNI site.
Targetig of pLNT-1 to YACs Following trasformation of YACs 410A3 and 37AB12 with linearised pLNT-1 (Figure 2A), again approximately 20-100 colonies per pg were observed on selective medium lacking lysine. Initial analysis of the recombinants by PFGE and hybridisation of the blot with neo showed for 410A3, 9/18 of the recombinants possessed neo and for 37AB12, 11/20 possessed neo (results not shown). Again the remainder not containing neo s resting from repair of the mutant are dtught to be Lys-2 gene in AB1380. TMe neo positive recombinants were analyzed fiuther by digeston with EcoRI (see Figure 2B). On hybridising the blot with neo as before either one or two bands were visible. Ihe lower band at 6kb repesents correct in at the EcoNI site of the right hand arm, whilst the upper band at 11.5kb represents the size of the vector following a multiple integration event. Six recombinants only psessed the lower band at 6kb representing correct integration at the EcoNI site in the right hand arm. On hybridising the same blot with pBR322 RHE, in the single integraton event there are two bands visible. The band at 6kb represents the same band as in the neo hybridisation, the band at 9.5kb represents the rest of the vector and some right hand arm sequences as expected (see Figure 2C). In the multiple
Targeting of pLNB-1 to YACs Yeast cells containing the 220kb YAC 410A3 were transformed with linearised pLNB-1 (Figure 3A). Initial analysis of 10 colonies growing on media lacking lysine, by hybridisation of a PFGE blot with neo showed that four of the clones had targeted to the correct size of the YAC whilst one had neo targeted to the Lys-2 gene present in the yeast and one had neo targeted at an increased size (data not shown). The odrer four of these clones had no neo present on hybridisation and presumably represented repair of the endogenous mutant Lys-2 gene (data not shown). Further PFGE analysis was carried out following digestion of the recombinants with Not. The pLNB-1 vector contains a Notd site which on integration into the YAC should introduce the site into the human DNA of the YAC. On hybridisation with neo three of the recombinants showed integration into the left hand arm presumably by homology of the ampicillin resistance gene present in the vector and in the left hand arm of the YAC (see Figure 3B track 2). One recombinant had multiple insertions at different locations in the human DNA (data not shown), one clone had a single integration event but at the lys-2 gene (data not shown) and one clone had a single integration event showing a single band hybridising at 190kb (Figure 3B track 1). Further
3A
-
ii
Cl 0
-f.f)
A'
'S
::
CO'
AM-
:?-
.s HI
NB-
;BK
Lett
*.* .;z * 'y'a-_a
a
S
*
M c.t NE: Arir: E
- OK-t
F%gur
\
_ . o,
3. Modification of YAC 410A3 using PLNB-1. 3A. The YAC modification vector pLNB-1 was digested with EcoRI to creat recombinogenic ends. The region of homology between fth vector and the YAC is shaded. 3B. Not I digested YAC DNA was blotted and hybridsed with various probes. Lane 1 yAC targete by insertion into the centre of the humian DNA, Lane 2 YAC tageted by insertion into the left hand arm of the yAC spuriously. Lane 3, undigested, tarete YAC showing the non functiona Lys-2 gene at limiting mobility and the targeted YAC at 230kb. 3C. The vector pLNB-1 inserts by recomination at the EcoRI site to give the targeted YAC as shown. The fragments obtained by hybridising the gel with the various probes are shown underneath the modified YAC diagram.
Nucleic Acids Research, Vol. 20, No. 12 2975 analysis of this clone by hybridisation of the blot with lys-2 showed two bands. The band at approximately 450kb represents the non-functional Lys-2 gene present in the yeast following NotI digestion and the band hybridising at 40kb represented the fragment from the other side of the NotI site (Figure 3C). On hybridisation with pBR322 LHE both the bands previously seen hybridised i.e. at 190kb and 40kb. Following repeat hybridisation with of the blot with pBR322 RHE a band of 40kb was visible (Figure 3B). From the data obtained it was possible to show the orientation of the insertion (see Figure 3C).
Fusion of 37AB12-pLNA-1 with V79 cells Approximately 1 in 104 to 105 G418 resistant V79 clones were obtained following fusion with 37AB12 pLNA-1. Genomic DNA was restricted with EcoRI, blotted and hybridised with pBR322, lys-2, CFTR exons 19, 23 and D9. Hybridisation with pBR322 showed multiple bands indicating the transfer of YAC arm sequences (Figure 4A). The band at 13kb was indicative of the transfer of the left hand arm sequences as shown previously (Figure IB) in the targeted YAC. To check if there was transfer of yeast sequences the blot was rehybridised with lys-2. As previously mentioned, the AB1380 strain possesses a non functional Lys-2 gene shown by a band at 15kb (Figure 4B). In four out of the five cell lines analyzed this yeast band was shown to be present suggesting that at least part, if not all of the yeast genome was transferred to the cell-lines. The other band hybridising with lys-2 at 13kb shows transfer of the left hand arm sequences from the YAC. On hybridising the blot with D9, CFTR exon 19 and exon 23 there were three bands hybridising at 4.0kb (D9), 2.2kb (CFTR 23) and at 2.4kb (CFTR 19) as shown in (Figure 4C). This confirms the transfer of CFTR sequences from the extreme left hand end of the YAC (D9) and the right hand end of the YAC(exon 19, 23). To confirm the presence of the central portion of the YAC SacII digests were carried out and the fragments resolved by PFGE. A band hybridising to a CFTR exon 13 probe of 2 10kb showed complete transfer of this region (Figure 4D).
4A
4B
DISCUSSION We have demonstrated that using the vectors described in this paper, it is possible to modify YACs in Saccharomyces cerivisiae (AB1380) to introduce a phosphoglycerate kinase driven neomycin resistance gene and the yeast selectable marker for aaminoadipidate reductase (Lys-2). Targeting of the linearised vectors into the YACs by homologous recombination has been shown to be highly specific for the left and right hand arm of the YAC allowing the introduction of one or more copies of the neo gene. However, the alu targeting vector has been found to be less specific, producing artefactual integrations into the left hand arm of the YAC and into the mutated Lys-2 gene of Saccharomyces cerivisiae. The alu targeting vector only targets tol5Obp on either side of the EcoRI site which on restriction digestion creates the recombinogenic ends. Since the vector is 11.5kb in length, there is scope for incorrect targeting. Furthermore the alu repeats within the YAC may not possess perfect sequence homology or correct orientation for targeting, resulting in few correct targeting events. Nevertheless we did obtain one out of ten clones which had integrated correctly at an alu repeat. The use of LI line repeat sequences has been found to work better than the alu repeat (7). Both the left hand and right hand targeting vectors integrate by insertion whereas the alu targeting vector can integrate by insertion or by replacement. From our data, it is apparent that the recombination event we observed was by insertion since there was no reduction in size of the YAC following targeting. Replacement can cause deletions in the YAC insert DNA by recombination between two repetitive alu elements which lie in close proximity, as has been observed in other clones we have analyzed (J.Riley, unpublished). A significant background of neo negative, lys positive recombinants (approximately 50%) was observed in all cases of modification. This is likely to be due to recombination repair of the mutant Lys-2 gene in the AB1380 host. This is easily screened for by hybridising filter lifts of lys positive clones with a neo probe to find the lys positive, neo positive colonies.
4C
4D
23.1 -
I
.
9.4
k- 23.1
--- 6.6
.4
---4.4
6.6 .4.4
l 1
2
pBR322
2 1
LYS-2
2
CFTR 23, 19 and D9
CFTR 13
Figure 4. Fusion of the modified CFTR YAC (37AB12-pLNA-1) with V79 cells. Two independent clones obtained after spheroplast fusion of 37AB12-pLNA-l to V79 cells (Chinese hamster fibroblasts). 4A. Lanes 1 and 2 represent EcoRI restricted DNA from the two cloned cell lines following blotting and hybridisation with pBR322. The band at 13kb is indicative of transfer of the left hand arm. 4B. Hybridisation of the two clones with lys-2 showed transfer of the 13kb pBR322 fragment as before and transfer of the 15kb non functional (AB1380) lys-2 gene. 4C. Hybridisation of the two clones with CFTR exons 23,19 and D9 showed the presence of all three fragments. 4D. Finally following SstII digestion of the two clones, fractionation by PFGE and hybridisation with CFTR exonl3 showed a band hybridising at 220kb, indicative of the transfer of the majority of the CFTR gene as a single DNA fragment.
2976 Nucleic Acids Research, Vol. 20, No. 12 The YACs used for these studies were 37AB12 which contained the whole of the CFTR gene and 410A3 containing DR4,B haplotype and were obtained from two independent libraries constructed using the vector pYAC-4 and were in the yeast host Saccharomyces cerivisiae AB1380. The efficient recombination systems in yeast can also be used to generate mutations by co-transfection of a modified mutation-containing fragment with a selection cassette. For CFTR this may include the deletion of amino acid 508 (Phenylalanine) (18), which is the most commonly found mutation in cystic fibrosis patients. Neo positives may be tested for the correct integration of the mutation by the use of allele specific PCR or Amplification Refractory Mutation System, ARMS technology (19). The introduction of YACs into cell-lines by fusion is already proving to be a useful tool for the study of gene function (2,3). It may in the foreseeable future prove useful in the construction of transgenic models of human disease via the Embryonic Stem cell route. The modified YAC 37AB12-pLNA-l described here was initially introduced into V79 Chinese hamster fibroblast cells using cell fusion. The presence of G418 resistant colonies showed that the PGK-neo functioned well in the cell lines. The results demonstrated that complete transfer of the YAC had occurred and that both the YAC arms were transferred, along with the yeast DNA. To test if the incoming yeast DNA integrates randomly at multiple sites within the yeast genome, experiments using fluorescent in situ hybridisation were carried out (J.Morten unpublished). These studies clearly indicated that although the site of integration was apparently random, most, if not all of the incoming yeast DNA goes into the Chinese hamster fibroblast chromosomes in one block. This suggests that the incoming yeast DNA is unlikely to disrupt a large number of host genes. These vectors will prove a useful tool in modifying YACs to incorporate a mammalian selectable marker, for studies on gene expression in cell lines, where a large gene is under the control of its own cis acting factors.
ACKNOWLEDGEMENTS We would like to thank Dr P.Philippsen for the gift of pDP-6 and Denis Le Paslier for the gift of YAC 410A3. We would also like to thank the Biographics department at ICI Pharmaceuticals for the photographs appearing in this manuscript.
REFERENCES 1. Burke, D.T. , Carle, G.F. and Olson, M.V. (1987). Science., 236, 806-812. 2. Huxley,C., Hagino, Y., Schlessinger,D. and Olson,M.V. (1991). Genomics, 9, 742-750. 3. Gnirke, A., Barnes, T.S., Patterson, D., Schild, D., Featherstone,T. and Olson, M.V. (1991). EMBO J., 10, 1629-1634. 4. Huxley, C. and Gnirke, A. (1991). Bioessays, 13, 545-549. 5. Pachnis, V., Pevny, L., Rothstein, R. and Constantini, F. (1990). Proc. Natl. Acad. Sci. USA., 87, 5109-5113. 6. Eliceiri, B., Labella, T., Hagino, Y., Srivastava, A., Schlessinger,D., Pilia,G., Palmieri, G. and D'Urso, M. (1991). Proc. Natl. Acad. Sci. USA.,
88, 2179-2183.
7. Srivastava A, K. and Schlessinger, D. (1991). Gene, 103, 53-59. 8. Pavan, W.J., Hieter, P. and Reeves, R.H. (1990). Mol.Cell.Biol., 10,
4163-4169. 9. Fleig, U.N., Pridmore, R.D. and Philippsen, P. (1986). Gene, 46, 237-245. 10. Adra, C.N., Boer, P.H. and McBurnay, M.W. (1987). Gene, 60, 65-74. 11. Molecular Cloning A Laboratory Manual 2nd Edition. Eds Sambrook,J., Fritsch,E.F. and Maniatis,T.(1989). Cold Spring Harbour Press. ppl.38, 1.39 and 1.42. 12. Anand, R., Ogilvie, D.J., Butler, R., Riley, J.H., Finniear, R.S., Powell,S., Smith,J.C and Markham, A.F. (1991). Genomics, 9, 124-130.
13. Anand, R., Riley, J.H., Butler, R., Smith, J.C. and Markham, A.F. (1990).
Nucl. Acid, Res.,18, 1951-1956. 14. Burgers, P. M. J. and Percival, K. J. (1987). Anal. Biochem., 163, 391-397. 15. Gel Electrophoresis of Nucleic Acids; A Practical Approach. Eds Rickwood, D. and Hames, B, D. IRL Press. pp 101-123. Anand, R. and Southern, E, M. (1989). 16. Estivill, X., McLean,C., Nunes, V., Casals, T., Gallano,P., Scambler, P. and Williamson, R. (1989). Am.J. Hum. Genet., 44, 704-710. 17. Feinberg, A,P. and Vogelstein, B. (1983). Anal. Biochem., 132, 6-13. 18. Riordan, J. R., Rommens, J. H., Kerem, B.S., Alon, N., Rozmahel, R.,
Grzelczak, Z., Zielenski, J., Lok, S., Plavsic, N., Chou, J.L., Drumm, M.L., lannuzzi, M.C., Collins, F.S., and Tsui, L.C. (1989). Science., 24s,
1066-1073. 19. Newton, C.R., Graham, A., Heptinstall, L.E., Powell, S.J., Summers, C., Kalsheker, N., Smith, J.C., and Markham, A,F. (1989). Nucl. Acid, Res., 17, 2593-2516.
