The combined presence of CD20 + B cells and PD-L1 + tumor-infiltrating lymphocytes in inflammatory breast cancer is prognostic of improved patient outcome.

PURPOSE: The purpose of the study was to evaluate protein expression of PD-L1 and CD20 as prognostic biomarkers of patient outcome in inflammatory breast cancer (IBC) samples. METHODS: PD-L1 and CD20 protein expression was measured by immunohistochemistry in 221 pretreatment IBC biopsies. PD-L1 was assessed in tumor cells (PD-L1+ tumor cells) and tumor stromal infiltrating lymphocytes (PD-L1+ TILs); CD20 was scored in tumor-infiltrating B cells. Kaplan-Meier curves and Cox proportional hazard models were used for survival analysis. RESULTS: PD-L1+ tumor cells, PD-L1+ TILs, and CD20+ TILs were found in 8%, 66%, and 62% of IBC, respectively. PD-L1+ tumor cells strongly correlated with high TILs, pathological complete response (pCR), CD20+ TILs, but marginally with breast cancer-specific survival (BCSS, P = 0.057). PD-L1+ TILs strongly correlated with high TILs, CD20+ TILs, and longer disease-free survival (DFS) in all IBC and in triple-negative (TN) IBC (P < 0.035). IBC and TN IBC patients with tumors containing both CD20+ TILs and PD-L1+ TILs (CD20+TILs/PD-L1+TILs) showed longer DFS and improved BCSS (P < 0.002) than patients lacking both, or those with either CD20+ TILs or PD-L1+ TILs alone. In multivariate analyses, CD20+TILs/PD-L1+TILs status was an independent prognostic factor for DFS in IBC (hazard ratio (HR): 0.53, 95% CI 0.37-0.77) and TN IBC (HR: 0.39 95% CI 0.17-0.88), and for BCSS in IBC (HR: 0.60 95% CI 0.43-0.85) and TN IBC (HR: 0.38 95% CI 0.17-0.83). CONCLUSION: CD20+TILs/PD-L1+TILs status represents an independent favorable prognostic factor in IBC and TN IBC, suggesting a critical role for B cells in antitumor immune responses. Anti-PD-1/PD-L1 and B cell-activating immunotherapies should be explored in these settings.

Fine-mapping of genes determining extrafusal fiber properties in murine soleus muscle.

Muscle fiber cross-sectional area (CSA) and proportion of different fiber types are important determinants of muscle function and overall metabolism. Genetic variation plays a substantial role in phenotypic variation of these traits; however, the underlying genes remain poorly understood. This study aimed to map quantitative trait loci (QTL) affecting differences in soleus muscle fiber traits between the LG/J and SM/J mouse strains. Fiber number, CSA, and proportion of oxidative type I fibers were assessed in the soleus of 334 genotyped female and male mice of the F34 generation of advanced intercross lines (AIL) derived from the LG/J and SM/J strains. To increase the QTL detection power, these data were combined with 94 soleus samples from the F2 intercross of the same strains. Transcriptome of the soleus muscle of LG/J and SM/J females was analyzed by microarray. Genome-wide association analysis mapped four QTL (genome-wide P < 0.05) affecting the properties of muscle fibers to chromosome 2, 3, 4, and 11. A 1.5-LOD QTL support interval ranged between 2.36 and 4.67 Mb. On the basis of the genomic sequence information and functional and transcriptome data, we identified candidate genes for each of these QTL. The combination of analyses in F2 and F34 AIL populations with transcriptome and genomic sequence data in the parental strains is an effective strategy for refining QTL and nomination of the candidate genes.

Constitutively activated PI3K accelerates tumor initiation and modifies histopathology of breast cancer.

The gene encoding phosphatidylinositol 3-kinase catalytic subunit α-isoform (PIK3CA, p110α) is frequently activated by mutation in human cancers. Based on detection in some breast cancer precursors, PIK3CA mutations have been proposed to have a role in tumor initiation. To investigate this hypothesis, we generated a novel mouse model with a Cre-recombinase regulated allele of p110α (myristoylated-p110α, myr-p110α) along with p53fl/fl deletion and KrasG12D also regulated by Cre-recombinase. After instillation of adenovirus-expressing Cre-recombinase into mammary ducts, we found that myr-p110α accelerated breast tumor initiation in a copy number-dependent manner. Breast tumors induced by p53fl/fl;KrasG12D with no or one copy of myr-p110α had predominantly sarcomatoid features, whereas two copies of myr-p110α resulted in tumors with a carcinoma phenotype. This novel model provides experimental support for importance of active p110α in breast tumor initiation, and shows that the amount of PI3K activity can affect the rate of tumor initiation and modify the histological phenotype of breast cancer.

In situ vaccination with cowpea mosaic virus nanoparticles suppresses metastatic cancer.

Nanotechnology has tremendous potential to contribute to cancer immunotherapy. The 'in situ vaccination' immunotherapy strategy directly manipulates identified tumours to overcome local tumour-mediated immunosuppression and subsequently stimulates systemic antitumour immunity to treat metastases. We show that inhalation of self-assembling virus-like nanoparticles from cowpea mosaic virus (CPMV) reduces established B16F10 lung melanoma and simultaneously generates potent systemic antitumour immunity against poorly immunogenic B16F10 in the skin. Full efficacy required Il-12, Ifn-γ, adaptive immunity and neutrophils. Inhaled CPMV nanoparticles were rapidly taken up by and activated neutrophils in the tumour microenvironment as an important part of the antitumour immune response. CPMV also exhibited clear treatment efficacy and systemic antitumour immunity in ovarian, colon, and breast tumour models in multiple anatomic locations. CPMV nanoparticles are stable, nontoxic, modifiable with drugs and antigens, and their nanomanufacture is highly scalable. These properties, combined with their inherent immunogenicity and demonstrated efficacy against a poorly immunogenic tumour, make CPMV an attractive and novel immunotherapy against metastatic cancer.

SU-E-I-81: Toward in Vivo Magnetic Spectroscopy of Brownian Motion.

PURPOSE: Magnetic spectroscopy of Brownian motion (MSB) has been used previously to measure temperature, viscosity, and cellular binding in vitro. The MSB signal - a ratio of the 5th to 3rd harmonic of the response from magnetic nanoparticles to an oscillating field - provides insight into particle microenvironment. These biosensing capabilities would be productive in vivo but until now were prevented by sensitivity limits. Our goal was to design and create a similar apparatus for work in vivo. In vivo spectroscopy is a viable precursor to imaging, and is essential for drug delivery or therapeutic methods like hyperthermia. METHODS: Coil geometries were modeled to optimize a uniform Helmholtz drive coil and imaging coil with maximal spatial resolution. The completed apparatus includes balancing and trim coils to zero out unwanted background fields. The coils were characterized and experiments were performed to verify consistency with previous in vitro experiments. Finally, as an in vivo experiment, we took MSB spectra on living mice with five week old melanomas injected with 200ug of 100nm starch coated nanoparticles. RESULTS: The drive coil is capable of sustaining 12.5mT fields up to 1.5kHz with a field variation of 3% throughout the sample volume. The pickup coil is frequency independent and has a vertical and horizontal range of 5mm and 10mm respectively before the MSB signal drops below 50%. The minimum sensitivity is 50-70µg of iron. MSB signal response to viscosity changes shows the same signatures as the in vitro apparatus. The in vivo data showed successful sensing of nanoparticles. We also saw the MSB signal decay with time showing the apparatus can detect changes in particle behavior due to interactions with biology. CONCLUSIONS: We achieved in vivo MSB and due to sufficient sensitivity we are motivated to further work in monitoring in vivo cellular uptake and viscosity.

Deoxyribonuclease IIalpha is required during the phagocytic phase of apoptosis and its loss causes perinatal lethality.

Deoxyribonuclease IIalpha (DNase IIalpha) is one of many endonucleases implicated in DNA digestion during apoptosis. We produced mice with targeted disruption of DNase IIalpha and defined its role in apoptosis. Mice deleted for DNase IIalpha die at birth with many tissues exhibiting large DNA-containing bodies that result from engulfed but undigested cell corpses. These DNA-containing bodies are pronounced in the liver where fetal definitive erythropoiesis occurs and extruded nuclei are degraded. They are found between the digits, where apoptosis occurs, and in many other regions of the embryo. Defects in the diaphragm appear to cause death of the mice due to asphyxiation. The DNA in these bodies contains 3'-hydroxyl ends and therefore stain positive in the TUNEL assay. In addition, numerous unengulfed TUNEL-positive cells are observed throughout the embryo. Apoptotic cells are normally cleared rapidly from a tissue; hence the persistence of the DNA-containing bodies and TUNEL-positive cells identifies sites where apoptosis occurs during development. These results demonstrate that DNase IIalpha is not required for the generation of the characteristic DNA fragmentation that occurs during apoptosis but is required for degrading DNA of dying cells and this function is necessary for proper fetal development.

Targeted disruption of the type 2 selenodeiodinase gene (DIO2) results in a phenotype of pituitary resistance to T4.

The type 2 deiodinase (D2), a selenoenzyme that catalyzes the conversion of T4 to T3 via 5'-deiodination, is expressed in the pituitary, brain, brown adipose tissue (BAT), and the reproductive tract. To examine the physiological role of this enzyme, a mouse strain lacking D2 activity was developed using homologous recombination. The targeting vector contained the Neo gene in place of a 2.6-kb segment of the Dio2 gene. This segment comprises 72% of the coding region and includes the TGA codon that codes for the selenocysteine located at the active site of the enzyme. Mice homologous for the targeted deletion [D2 knockout (D2KO)] had no gross phenotypic abnormalities, and development and reproductive function appeared normal, except for mild growth retardation (9%) in males. No D2 activity was observed in any tissue in D2KO mice under basal conditions, or under those that normally induce this enzyme such as cold-exposure (BAT) or hypothyroidism (brain, BAT, and pituitary gland). Furthermore, no D2 activity was present in cultured astrocytes, nor could it be induced by treatment of the cells with forskolin. Although D2 mRNA transcripts were detected in BAT RNA obtained from cold-exposed wild-type (WT) mice, none was detected in BAT RNA from comparably-treated D2KO mice. Levels of D1 in the liver, thyroid, and pituitary were the same in WT and D2KO animals, whereas D3 activity in D2KO cerebrum was twice that in WT cerebrum. Serum T3 levels were comparable in adult WT and D2KO mice. However, serum T4 and TSH levels were both elevated significantly (40% and 100%, respectively) in the D2KO mice, suggesting that the pituitary gland of the D2KO mouse is resistant to the feedback effect of plasma T4. This view was substantiated by the finding that serum TSH levels in hypothyroid WT mice were suppressed by administration of either T4 or T3, but only T3 was effective in the D2KO mouse. The data also suggest that the clearance of T4 from plasma was reduced in the D2KO mouse. In summary, targeted inactivation of the selenodeiodinase Dio2 gene results in the complete loss of D2 activity in all tissues examined. The increased serum levels of T4 and TSH observed in D2KO animals demonstrate that the D2 is of critical importance in the feedback regulation of TSH secretion.

Targeted deletion of 5'HS1 and 5'HS4 of the beta-globin locus control region reveals additive activity of the DNaseI hypersensitive sites.

The mammalian beta-globin locus is a multigenic, developmentally regulated, tissue-specific locus from which gene expression is regulated by a distal regulatory region, the locus control region (LCR). The functional mechanism by which the beta-globin LCR stimulates transcription of the linked beta-like globin genes remains unknown. The LCR is composed of a series of 5 DNaseI hypersensitive sites (5'HSs) that form in the nucleus of erythroid precursors. These HSs are conserved among mammals, bind transcription factors that also bind to other parts of the locus, and compose the functional components of the LCR. To test the hypothesis that individual HSs have unique properties, homologous recombination was used to construct 5 lines of mice with individual deletions of each of the 5'HSs of the endogenous murine beta-globin LCR. Here it is reported that deletion of 5'HS1 reduces expression of the linked genes by up to 24%, while deletion of 5'HS4 leads to reductions of up to 27%. These deletions do not perturb the normal stage-specific expression of genes from this multigenic locus. In conjunction with previous studies of deletions of the other HSs and studies of deletion of the entire LCR, it is concluded that (1) none of the 5'HSs is essential for nearly normal expression; (2) none of the HSs is required for proper developmental expression; and (3) the HSs do not appear to synergize either structurally or functionally, but rather form independently and appear to contribute additively to the overall expression from the locus.

Position effects are influenced by the orientation of a transgene with respect to flanking chromatin.

We have inserted two expression cassettes at tagged reference chromosomal sites by using recombinase-mediated cassette exchange in mammalian cells. The three sites of integration displayed either stable or silencing position effects that were dominant over the different enhancers present in the cassettes. These position effects were strongly dependent on the orientation of the construct within the locus, with one orientation being permissive for expression and the other being nonpermissive. Orientation-specific silencing, which was observed at two of the three site tested, was associated with hypermethylation but not with changes in chromatin structure, as judged by DNase I hypersensitivity assays. Using CRE recombinase, we were able to switch in vivo the orientation of the transgenes from the permissive to the nonpermissive orientation and vice versa. Switching from the permissive to the nonpermissive orientation led to silencing, but switching from the nonpermissive to the permissive orientation did not lead to reactivation of the transgene. Instead, transgene expression occurred dynamically by transcriptional oscillations, with 10 to 20% of the cells expressing at any given time. This result suggested that the cassette had been imprinted (epigenetically tagged) while it was in the nonpermissive orientation. Methylation analysis revealed that the methylation state of the inverted cassettes resembled that of silenced cassettes except that the enhancer had selectively lost some of its methylation. Sorting of the expressing and nonexpressing cell populations provided evidence that the transcriptional oscillations of the epigenetically tagged cassette are associated with changes in the methylation status of regulatory elements in the transgene. This suggests that transgene methylation is more dynamic than was previously assumed.

Independent formation of DnaseI hypersensitive sites in the murine beta-globin locus control region.

Mammalian beta-globin loci are composed of multiple orthologous genes whose expression is erythroid specific and developmentally regulated. The expression of these genes both from the endogenous locus and from transgenes is strongly influenced by a linked 15-kilobase region of clustered DNaseI hypersensitive sites (HSs) known as the locus control region (LCR). The LCR encompasses 5 major HSs, each of which is highly homologous among humans, mice, and other mammals. To analyze the function of individual HSs in the endogenous murine beta-globin LCR, we have used homologous recombination in embryonic stem cells to produce 5 mouse lines, each of which is deficient for 1 of these major HSs. In this report, we demonstrate that deletion of the conserved region of 5'HS 1, 2, 3, 4, or 5/6 abolishes HS formation at the deletion site but has no influence on the formation of the remaining HSs in the LCR. Therefore, in the endogenous murine locus, there is no dominant or initiating site whose formation must precede the formation of the other HSs. This is consistent with the idea that HSs form autonomously. We discuss the implications of these findings for current models of beta-globin regulation.

Beta-globin YAC transgenes exhibit uniform expression levels but position effect variegation in mice.

Expression of a construct integrated at different genomic locations often varies because of position effects that have been subcategorized as stable (decreased level of expression) and variegating (decreased proportion of expressing cells). It is well established that locus control regions (LCRs) generally overcome position effects in transgenes. However, whether stable and variegated position effects are equally overcome by an intact LCR has not been determined. We report that single-copy yeast artificial chromosome transgenes containing an unmodified human beta -globin locus were not subject to detectable stable position effects but did undergo mild to severe variegating position effects at three of the four non-centromeric integration sites tested. We also find that, at a given integration site, the distance and the orientation of the LCR relative to the regulated gene contributes to the likelihood of variegating position effects, and can affect the magnitude of its transcriptional enhancement. DNase I hypersensitive site (HSS) formation varies with the proportion of expressing cells, not the level of gene expression, suggesting that silencing of the transgene is associated with a lack of HSS formation in the LCR region. We conclude that transcriptional enhancement and variegating position effects are caused by fundamentally different but inter-dependent mechanisms.

Deletions within the mouse beta-globin locus control region preferentially reduce beta(min) globin gene expression.

The mouse beta-globin gene cluster is regulated, at least in part, by a locus control region (LCR) composed of several developmentally stable DNase I hypersensitive sites located upstream of the genes. In this report, we examine the level of expression of the beta(min) and beta(maj) genes in adult mice in which HS2, HS3, or HS5,6 has been either deleted or replaced by a selectable marker via homologous recombination in ES cells. Primer extension analysis of RNA extracted from circulating reticulocytes and HPLC analysis of globin chains from peripheral red blood cells revealed that all mutations that reduce the overall output of the locus preferentially decrease beta(min) expression over beta(maj). The implications of these findings for the mechanism by which the LCR controls expression of the beta(maj) and beta(min) promoters are discussed.

To be or not to be active: the stochastic nature of enhancer action.

Transcriptional enhancers are traditionally considered to regulate the rate at which a linked promoter transcribes mRNA, but recent experiments suggest a reevaluation of this model is necessary. Single-cell assays of transgenes reveal that enhancers increase the probability that a reporter gene will be active, but have little or no effect on the transcription rate once a gene has been activated. These results raise the question of how enhancers affect gene expression in their native contexts. A simple interpretation is that enhancers act in a stochastic fashion to increase the probability that a regulated gene will be transcribed; such a model is compatible with programs of cell differentiation in which multiple similar cells subject to similar environmental stimuli do not respond uniformly.

Site-specific chromosomal integration in mammalian cells: highly efficient CRE recombinase-mediated cassette exchange.

Expression of experimental constructs in mammalian cells or transgenic animals is difficult to control because it is markedly influenced by position effects. This has limited both the analysis of cis -DNA regulatory elements for transcription and replication, and the physiological analysis of proteins expressed from transgenes. We report here two new methods based on the concept of recombinase-mediated cassette exchange (RMCE) to perform site-specific chromosomal integration. The first method permits the exchange of a negative selectable marker pre-localized on the chromosome with a transgene via a CRE-mediated double recombination between inverted Lox sites. Integration efficiency is close to 100 % of negatively selected mouse erythroleukemia cells and ranges from 10 to 50 % in embryonic stem cells. The second method allows RMCE with no selection at all except for cells that have taken up plasmid transiently. While less efficient, this technique permits novel experimental approaches. We find that integration of a transgene at a given genomic site leads to reproducible expression. RMCE should be useful to develop artificial genetic loci that impart specific and reproducible regulation of transgenes in higher eukaryotes. This should facilitate the analysis of cis -regulatory DNA elements governing expression and position effects, improve our control over the physiological effects of transgenes, and accelerate the development of animal models for complex human diseases.

Analysis of mammalian cis-regulatory DNA elements by homologous recombination.

The use of homologous recombination to modify and thereby functionally analyze cis-regulatory DNA elements in mammalian cells has become an important approach in mammalian gene expression research. We have emphasized the necessity of designing a system that allows the removal of selectable markers used in targeting and facilitates the further modification of the region under study. To perform these tasks, we presently favor making an initial HR-mediated replacement of the entire element under study with an active positive selectable marker in combination with either an inactive second positive selectable marker or an active negative selectable marker. The plug and socket system, in which an inactive selectable marker is complemented by HR, is the most dependable and well-characterized option for making secondary modifications. However, the double-replacement system has certain advantages, and the recently developed RMCE approach, which allows replacement of a negative selectable marker by site-specific recombinase-mediated insertion without using a positive selectable marker, will likely prove very valuable in future experiments. Each of the systems, or combinations thereof, should be considered in light of the specifics of any given experiment to select the most appropriate option. Although the emphasis of this article has been the analysis of cis-acting regulatory elements involved in transcription, these same approaches can be used to analyze other regulatory elements (e.g., origins of replication) and to make multiple subtle mutations in polypeptides.

The chicken beta-globin 5'HS4 boundary element blocks enhancer-mediated suppression of silencing.

A constitutive DNase I-hypersensitive site 5' of the chicken beta-globin locus, termed 5'HS4 or cHS4, has been shown to insulate a promoter from the effect of an upstream enhancer and to reduce position effects on mini-white expression in Drosophila cells; on the basis of these findings, it has been designated a chromatin insulator. We have examined the effect of the cHS4 insulator in a system that assays both the level of gene expression and the rate of transcriptional silencing. Because transgenes flanked by insulator elements are shielded from position effects in Drosophila cells, we tested the ability of cHS4 to protect transgenes from position effects in mammalian cells. Flanking of an expression vector with the cHS4 insulator in a colony assay did not increase the number of G418-resistant colonies. Using lox/cre-based recombinase-mediated cassette exchange to control integration position, we studied the effect of cHS4 on the silencing of an integrated beta-geo reporter at three genomic sites in K562 erythroleukemia cells. In this assay, enhancers act to suppress silencing but do not increase expression levels. While cHS4 blocked enhancement at each integration site, the strength of the effect varied from site to site. Furthermore, at some sites, cHS4 inhibited the enhancer effect either when placed between the enhancer and the promoter or when placed upstream of the enhancer. These results suggest that the activity of cHS4 is not dominant in all contexts and is unlikely to prevent silencing at all genomic integration sites.

Description and targeted deletion of 5' hypersensitive site 5 and 6 of the mouse beta-globin locus control region.

The most upstream hypersensitive site (HS) of the beta-globin locus control region (LCR) in humans (5' HS 5) and chickens (5' HS 4) can act as an insulating element in some gain of function assays and may demarcate a beta-globin domain. We have mapped the most upstream HSs of the mouse beta-globin LCR and sequenced this region. We find that mice have a region homologous to human 5' HS 5 that is associated with a minor HS. In addition we map a unique HS upstream of 5' HS 5 and refer to this novel site as mouse 5' HS 6. We have also generated mice containing a targeted deletion of the region containing 5' HS 5 and 6. We find that after excision of the selectable marker in vivo, deletion of 5' HS 5 and 6 has a minimal effect on transcription and does not prevent formation of the remaining LCR HSs. Taken together these findings suggest that the most upstream HSs of the mouse beta-globin LCR are not necessary for maintaining the beta-globin locus in an active configuration or to protect it from a surrounding repressive chromatin environment.

The beta-globin LCR is not necessary for an open chromatin structure or developmentally regulated transcription of the native mouse beta-globin locus.

The murine beta-globin locus control region (LCR) was deleted from its native chromosomal location. The approximately 25 kb deletion eliminates all sequences and structures homologous to those defined as the human LCR. In differentiated ES cells and erythroleukemia cells containing the LCR-deleted chromosome, DNasel sensitivity of the beta-globin domain is established and maintained, developmental regulation of the locus is intact, and beta-like globin RNA levels are reduced 5%-25% of normal. Thus, in the native murine beta-globin locus, the LCR is necessary for normal levels of transcription, but other elements are sufficient to establish the open chromatin structure, transcription, and developmental specificity of the locus. These findings suggest a contributory rather than dominant function for the LCR in its native location.

Reduced beta-globin gene expression in adult mice containing deletions of locus control region 5' HS-2 or 5' HS-3.

To gain insights into the functions of individual DNA'se hypersensitive sites within the beta globin locus control region (LCR), we deleted the endogenous 5' HS-2 and HS-3 regions from the mouse germline using homologous recombination techniques. We demonstrated that the deletion of either murine 5' HS-2 or 5' HS-3 reduced the expression of the embryonic epsilon y and beta h1 globin genes minimally in yolk sac-derived erythrocytes, but that both knockouts reduced the output of the adult beta (beta-Major + beta-Minor) globin genes by approximately 30% in adult erythrocytes. When the selectable marker PGK-Neo cassette was retained within either the HS-2 or HS-3 region, a much more severe reduction in globin gene expression was observed at all developmental stages. PGK-Neo was shown to be expressed in an erythroid-specific fashion when it was retained in the HS-3 position. These results show that neither 5' HS-2 nor HS-3 is required for the activity of embryonic globin genes, nor are these sites required for correct developmental switching. However, each site is required for approximately 30% of the total LCR activity associated with adult beta-globin gene expression in adult red blood cells. Each site therefore contains some non-redundant information that contributes to adult globin gene function.

DNA cassette exchange in ES cells mediated by Flp recombinase: an efficient strategy for repeated modification of tagged loci by marker-free constructs.

The repeated modification of a genomic locus is a technically demanding but powerful strategy to analyze the function of a particular gene product or the role of cis-regulatory DNA elements in mammalian cells. The initial step is "tagging" a site with a selectable marker which is done by homologous recombination (HR) to modify a known locus or by random integration to study cis-regulatory elements at a reproducibly accessible genomic location. The tag is then used to target the construct of choice during an exchange step. Presented here is a novel technique in which the exchange is independent of HR and does not introduce vector sequences. It relies on our previous studies on the replacement of DNA cassettes by FLP-recombinase, whereby some common limitations can be overcome. To this end, the tag, a hygtk positive/negative selection marker, is integrated into the genome of embryonic stem (ES) cells. This marker is flanked by a wild-type Flp-recognition target (FRT) site on one end and by a modified heterospecific FRT site on the other. Successful Flp-mediated replacement of the hygtk cassette is enriched by ganciclovir (GANC) selection for cells that lack the encoded fusion protein. Thereby, the hygtk gene can be exchanged for virtually any sequence in a single efficient step without the need of introducing a positive selectable marker. The system can hence be used to analyze the function of either a gene product or regulatory sequences in ES cells or the transgenic mice derived thereof.