Modeling for Lesch-Nyhan Disease by Gene Targeting in Human Embryonic Stem Cells
http://www.100md.com
《干细胞学杂志》
Department of Genetics, Silberman Institute of Life Science, The Hebrew University, Jerusalem, Israel
Key Words. Human ES cells ? Genetic diseases ? Lesch-Nyhan syndrome ? Homologous recombination
Correspondence: Nissim Benvenisty M.D., Ph.D., Hebrew University Department of Genetics, Institute of Life Sciences, Givat-Ram, Jerusalem 91904 Israel. Telephone: 972-2-6586774; Fax: 972-2-6584972; e-mail: nissimb@mail.ls.huji.ac.il
ABSTRACT
Human embryonic stem (ES) cells are pluripotent cells derived from blastocyst-stage embryos. It has been suggested that these cells should play a major role in transplantation medicine and be able to advance our knowledge in human embryology. We propose that these cells should also play a vital role in the creation of models of human disorders. This aspect would be most valuable where animal models failed to faithfully recapitulate the human phenotype. Lesch-Nyhan disease is caused by a mutation in the HPRT1 gene that triggers an overproduction of uric acid, causing gout-like symptoms and urinary stones, in addition to neurological disorders. Due to biochemical differences between humans and rodents, a mouse lacking the HPRT expression will fail to accumulate uric acid. In this research we demonstrate a model for Lesch-Nyhan disease by mutating the HPRT1 gene in human ES cells using homologous recombination. We have verified the mutation in the HPRT1 allele at the DNA and RNA levels. By using selection media, we show that HPRT1 activity is abolished in the mutant cells, and the HPRT1–cells show a higher rate of uric acid accumulation than the wild-type cells. Therefore, these cells recapitulate to some extent the characteristics of Lesch-Nyhan syndrome and can help researchers further investigate this genetic disease and analyze drugs that will prevent the onset of its symptoms. We therefore suggest that human diseases may be modeled using human ES cells.
INTRODUCTION
Human embryonic stem (ES) cells are pluripotent cells derived from the inner cell mass of blastocyst-stage embryos . These cells have two main properties: self-renewal capacity and pluripotent differentiation potential. The pluripotency of the cells has been shown reproducibly in vitro, where human ES cells produce embryonic bodies (EBs) that comprise the three embryonic germ layers . These cells also have the ability for in vivo differentiation as they form differentiated tumors upon injection into immunodeficient severe combined immunodeficiency syndrome/Beige mice . These tumors contain cells from all three germ layers.
These properties of the human ES cells give them a major role in the study of early human development and in medical research because of their potential role in transplantation therapy. However, human ES cells also enable us to generate models for human diseases. In this work we created a model for Lesch-Nyhan disease by inducing a mutation in human ES cells. Lesch-Nyhan syndrome is a severe genetic disease that is caused by the malfunction of the hypoxanthine phosphoribosyltransferase (HPRT1) gene . One main clinical feature of this disease is hyperuricemia that leads to fatal uric acid urinary stones and a gout-like syndrome. In addition, the disease exhibits characteristic neurological symptoms such as mental retardation, spastic cerebral palsy, choreoathetosis, and self-destructive biting of fingers and lips. It is still not clear how the deficiency in HPRT1 activity leads to these neurological symptoms, although it is documented that dopaminergic neurons are damaged in affected patients . The biochemical features leading to overproduction of uric acid are due to an inability to convert hypoxanthine into inosine 5'-monophosphate and the conversion of hypoxanthine into uric acid. In rodents, uric acid does not accumulate because of the activity of the urate oxidase (UOX) enzyme that converts uric acid into allantoin . In addition, no clear neurobehavioral defects have been observed in mouse models that exhibit HPRT1 deficiency. Thus, no good model for Lesch-Nyhan syndrome has been achieved in rodents .
Homologous recombination recently has been described as a method to induce mutation in human ES cells . In order to create a model for Lesch-Nyhan disease in human cells, we have targeted HPRT1 in human ES cells by homologous recombination, using a replacement-type vector that is based on nonisogenic DNA. The mutation in the HPRT1 gene by homologous recombination was confirmed at the DNA and RNA levels and by using selection media to demonstrate an absence in HPRT1 activity. The mutant cells produced significantly more uric acid than wild-type cells do. We believe that these cells will enable further research on Lesch-Nyhan disease, especially in aspects where using mouse models was limited. In addition, our analysis shows that homologous recombination in human ES cells can be achieved with a nonisogenic construct transfected with a cationic reagent.
MATERIALS AND METHODS
Cell Culture
Male human ES cells (H13) were cultured as described by Itskovitz-Eldor et al. using media consisting of 85% KnockoutTMD-MEM (Gibco-Invitrogen Corporation, Carls-bad, CA), 15% KnockoutTMSR—a serum-free formulation (Gibco-Invitrogen Corporation), 1 mM glutamine (Gibco-Invitrogen Corporation, St. Louis), 0.1 mM ?-mercaptoethanol (Sigma), 1% nonessential amino acids stock (Gibco-Invitrogen Corporation), 4 ng/ml basic fibroblast growth factor (bFGF; Gibco-Invitrogen Corporation), and ITS 100X (insulin transferrin selenium), 1:200 dilution (Gibco-Invitrogen Corporation). EBs were formed by transferring 2–4 x 106 ES cells using trypsin/EDTA (0.1%/1 mM) to plastic Petri dishes (Miniplats, Ein-Shemer, Israel) in order to allow their aggregation. The EBs were grown in the same medium as ES cells but without bFGF.
Transfection and Establishment of Mutant Cell Lines
PHPRThyg , a plasmid that contains an insertion of the hygromycin resistance gene into an 8.5-kb fragment of exons II and III of the HPRT1 gene, was introduced into human ES cells using ExGen 500 (Fermentas, Hanover, MD) as described by Eiget et al . More specifically, 106 cells were transfected with 3–4 μg of plasmid liner DNA, centrifuged, and incubated for 30 minutes with the transfection reagent. Twenty-four hours later cells were replated and after 4 days were selected with either 6-thioguanine (6TG, 1 μg/ml; Sigma), hygromycin (100 μg/ml; Roche, Basel, Switzerland), or both.
Selection Media and Cell Proliferation
Selection media were added to wild-type and mutant cells to test for the activity of the purine salvage pathway. Thus, human ES cells were grown with 6TG (1 μg/ml) inhibiting the growth of cells that express HPRT1. Alternatively, the cells were incubated with HAT (1 x 10–4 M hypoxanthine, 4 x 10–7 M aminopterin, 1.6 x 10–5 M thymidine; Sigma), which inhibits the de novo pathway of purine synthesis and thus prevents the growth of cells that do not express HPRT1. One to three days after adding the selection media, we analyzed cell densities by fixating cells with 2.5% glutaraldehyde (Merck, Whitehouse Station, NJ) to the tissue culture plates. The plates were stained with methylene blue (Sigma) and dissolved in 0.1 M boric acid (pH = 8.5). We extracted color with 0.1 N HCl and read emission at 650 nm. Because color density represents cell number, we prepared growth charts based on color emission.
PCR
Total DNA was extracted using EZ-DNA kit (Biological Industries, Beit Haemek, Israel). DNA samples were subjected to polymerase chain reaction (PCR) amplification using the following primers: I: GCAAGTACTCA-GAACAGCTGC; II: TGATGAACCAGGTTATGACC; III: CATCGAAGCTGAAAGCACCAG. The combination of primer I and II should recognize only the wild-type HPRT1 sequence, while the combination of II and III should identity only the mutant sequence (Fig. 1 caption).
Figure 1. Homologous recombination of the HPRT1 gene. (A): Strategy for homologous recombination of HPRT1 and expected polymerase chain reaction (PCR) products from the wild-type (wt) human embryonic stem (ES) cells and from clones mutated in HPRT1 by homologous recombination. I: A specifice primer for the wt cells; II: a primer that is common to the wt and mutant cells; III: a specific primer for the mutant cells. Shown, at the bottom of the scheme (*), the distribution of the single nucleotide polymorphisms along the recombined HPRT1 sequence. (B): PCR analysis of HPRT1 sequences in wt and in mutant ES cells. DNA was isolated from the cells and then subjected to PCR using the primers mentioned above. (C): Reverse transcriptase PCR analysis of wt and mutant cells; cDNA was preformed from mutant and wt RNA and then subjected to PCR using a set of primers that are located in exons I and V of the HPRT1 gene and are expected to identify only the wt HPRT1 mRNA sequence. Note that the HPRT1–clones lack expression of the wt transcript. Primers that amplify the housekeeping gene GAPDH were used as a control.
RNA Isolation and RT-PCR
Total RNA from ES cells was extracted as described by Chomczynski and Sacchi using guanidine thiocyanate followed by phenol-chloroform and isopropanol precipitation. EBs were subjected to homogenization by polytron and then extracted for total RNA by RNeasy? Mini Kit (Qiagen, Valencia, CA). We synthesized cDNA from 1 μg of total RNA using random hexamer (pd6) as primer (Pharmacia Biotech, New Jersey) and M-MLV reverse transcriptase (Gibco-Invitrogen Corporation). We subjected cDNA samples to PCR amplification with the primers shown in Table 1. For each gene, the DNA primers were derived from different exons to ensure that the PCR product represents the specific mRNA transcript and not genomic DNA.
Table 1. Primer sequences for reverse transcription polymerase chain reaction
Quantitation of Uric Acid Production
Mutant and wild-type human ES cells were grown in 4-ml media for 48 hours in the presence or absence of 100 μM hypoxanthine. The medium was concentrated to 1 ml by SpeedVac (Eppendorf, Hamburg, Germany). We assessed the level of uric acid using liquid chromatography.
RESULTS
In order to generate a model for Lesch-Nyhan disease, we decided to mutate the HPRT1 gene in human ES cells. HPRT1 is X-linked, and thus a targeted mutation of the only allele in male human ES cells may lead to biochemical features observed in Lesch-Nyhan patients. We targeted HPRT1 in human ES cells by homologous recombination using a replacement-type vector. This vector contains 8.5 kb of non-isogenic exonic and intronic sequences of the HPRT1 gene, where the hygromycin resistance marker is placed within exon II (Fig. 1A) . According to the National Center for Biotechnology Information (NCBI) database, this sequence contains 12 single nucleotide polymorphisms (SNPs) that are distributed along the entire sequence (Fig. 1A). Homologous recombination of this plasmid leads to HPRT1 inactivation, resulting in resistance to both hygromycin and 6TG, a nucleotide analog that inhibits growth of HPRT1-expressing cells. Out of more than 108 transfected cells, two human ES cell clones that were resistant to both drugs were isolated following ExGen 500–mediated transfection.
We confirmed the presence of the targeted HPRT1 insertion expected in these clones by PCR (Fig. 1B). In addition, we confirmed by reverse transcriptase PCR (RT-PCR) analysis that the mutated cells do not express the wild-type HPRT1 transcript (Fig. 1C). In order to prove that the HPRT1 gene is not activated in the mutant cells, we compared the proliferation rate of wild-type and mutant cells under the selection of 6TG and HAT. HAT medium inhibits the de novo pathway of purine synthesis and prevents growth of cells that lack an active salvage pathway. The mutant cells proliferate in the presence of 6TG and die in HAT medium while wild-type cells behave in an opposite fashion (Fig. 2). These results confirmed our ability to perform insertion mutagenesis by homologous recombination using a cationic reagent.
Figure 2. Analysis of HPRT1 activity using selection media. (A): In order to examine the activity of HPRT1, we used a set of selection media. One medium contains 6-thioguanine (6TG), which prevents growth of all cells in which the purine salvage pathway is active, while the second is the hypoxanthine-aminopterin-thymidine (HAT) medium, which prevents growth of cells that do not have an active purine salvage pathway. Growth curves of wild-type (wt) and mutant cells in the different selection media are shown. Note that the wt cells stop growing in the presence of 6TG, while the mutant cells stop growing in the presence of HAT. (B): Example of mutated cells growing in the different selection media and stained with methylene blue. Abbreviation: O.D., optical density.
Since one of the hallmarks of Lesch-Nyhan syndrome is the accumulation of uric acid, we have analyzed the levels of uric acid in our targeted cells grown in medium with and without the addition of hypoxanthine. Indeed, uric acid significantly accumulates more in mutant cells than in wild-type cells (Fig. 3). Allopurinol, a drug that affects uric acid accumulation in humans, also dramatically reduced uric acid accumulation in the cells (data not shown). Uric acid fails to accumulate in rodent cells mutated in HPRT since these cells express UOX, an enzyme that converts uric acid into allantoin (Fig. 3). Thus, HPRT1-deficient human ES cells recapitulate the biochemical phenotype observed in Lesch-Nyhan cells, while rodent cells failed to do so.
Figure 3. Uric acid production in HPRT1 cells. (A): Schematic pathway of hypoxanthine metabolism into uric acid and inosine 5'-monophosphate (IMP). Note the degradation of uric acid into allantoin occurs only in rodents and not in humans. (B): Uric acid levels (μM) in wild-type and in HPRT1-deficient human embryonic stem cells. Note that the mutant cells accumulate significantly more uric acid with or without the addition of 100 μM hypoxanthine.
The neurological symptoms of Lesch-Nyhan patients have been related to a malfunction in dopaminergic cells . Although it is yet not possible to achieve fully differentiated dopaminergic neurons that express tyrosine hydroxylase from human ES cells , we have shown that like wild-type cells, spontaneously differentiated mutant cells express a key dopaminergic enzyme—dopamine decarboxylase (DDC; Fig. 4).
Figure 4. Analysis of a dopaminergic marker in HPRT–cells. RNA was isolated from 21-day-old embryonic bodies (EBs) and was subjected to reverse transcriptase polymerase chain reaction using the primers that identify either a dopaminergic marker (dopamine decarboxylase ), an embryonic stem cell marker (OCT4), an endoderm differentiation marker (A-feto-protein ), or a housekeeping marker (GAPDH).
DISCUSSION
Models for human disorders are important tools in analyzing the phenotype and potential therapies for genetic diseases. Several approaches were taken in order to achieve such models both in vivo and in vitro. The creation of mutant mice using the homologous recombination technique was vital in generating animal models for many human diseases. However, morphological, developmental, and genetic differences between humans and mice posed significant limitations for the use of these mutant animals as models for human disorders. In many cases, the mouse phenotype largely differs from those observed in humans . In addition to in vivo animal models, researchers may study in vitro human systems to model human disorders. One such approach is the generation of primary cultures from patients’ cells. This methodology is limited due to the specific range of tissues from which cells can be achieved and the short life span of these primary cultures. In addition, these systems rely on mutations that occurred naturally in humans in a specific genetic background. A way of overcoming this limitation is to induce mutations in human cell lines by choice. Thus, homologous recombination has been exercised on various cell lines mutating mostly genes involved in oncogenesis and senescence . However, the mutations were induced in transformed cells, and in each case, the phenotype was studied in already differentiated, committed cell lines.
Human ES cells are nontransformed cells with the capacities of self-renewal and pluripotency, allowing them to differentiate into a wide range of cell types. So far the unique features of these cells were mainly used in the search for new sources of cell transplantation. However, via genetic manipulation, human ES cells can also be used to model human diseases. This is especially important where the animal in vivo models have failed to fully recapitulate the phenotype in the human disease.
A good example for such a case is Lesch-Nyhan syndrome. This syndrome is caused by accumulation of uric acid due to a mutation in the HPRT1 enzyme . Lesch-Nyhan patients accumulate uric acid and thus suffer from renal failure and a gout-like syndrome, in addition to neurobehavioral symptoms related to a malfunction of the dopaminergic neurons. Uric acid fails to accumulate in mice lacking HPRT activity since they express UOX, the enzyme that converts uric acid into allantoin (Fig. 3) . Thus, it is difficult, if not impossible, to create a faithful model for Lesch-Nyhan syndrome in mice. In our study, we have shown that directed mutagenesis in the HPRT1 gene in human ES cells results in high accumulation of uric acid compared to wild-type cells. Thus, we were able to generate a model for the biochemical phenotype of Lesch-Nyhan syndrome in a culture system. In our mutant cells, the drug allopurinol leads to a decrease in uric acid levels. Therefore, this system may be used, in the future, in the search for new medications for Lesch-Nyhan patients.
Since a significant phenotype in Lesch-Nyhan patients is related to dopaminergic cells, a major challenge is to find a way to induce complete dopaminergic differentiation in human ES cells. Achievement of this goal will enable us to examine biochemical and electrophysiological features in wild-type and mutant dopaminergic cells. Establishment of an in vitro system that mimics the development of undifferentiated HPRT1–cells to dopaminergic neurons may give an opportunity, which does not exist in any other models, to study the developmental processes that lead to the malfunction of the dopaminergic neurons in Lesch-Nyhan syndrome. This system should also allow a better understanding of these patients’neurological symptoms. Although at present conditions in which to differentiate human ES cells to dopaminergic cells are still unknown , the fact that the genetic manipulation of the cells did not prevent their partial differentiation into DDC-expressing cells encourages additional work with these mutant cells in order to further study the neurological phenotype of Lesch-Nyhan syndrome.
In our analysis, we have recombined sequences that were derived from separate human cell lines. A number of studies in mouse ES cells have shown that the use of nonisogenic sequences decreases the efficiency of homologous recombination by up to 20-fold . However, it has been suggested that the efficiency of gene targeting in human cells is the same for isogenic and nonisogenic sequences . In this study, the efficiency of nonhomologous recombination as determined by resistance to an antibiotic selection marker is about 10–6 transfected cells, while the efficiency of homologous recombination is about 10–8 transfected cells. Hence, the ratio of homologous recombination versus nonhomologous recombination in our system is estimated to be about 1%, which is only somewhat lower than that of 2% as recently reported for gene targeting with isogenic sequences by electroporation in human ES cells . Thus, our results support the notion presented by Sedivy et al. that homologous recombination in humans is efficient even with non-isogenic DNA. One possible explanation for the differences between humans and mice, in relation to the efficiency of nonisogenic homologous recombination, is that the sequence divergence between two human cell lines is much smaller than the difference between two different mouse strains. According to the NCBI database, there are 12 SNPs in the 8.5-kb sequences of HPRT1 used in our study for targeted recombination (1 SNP per 0.7 kb). This figure reflects the normal distribution of SNPs in the human genome among nonisogenic sequences . However, these differences are much smaller than those that appear in the sequences of different mice strains . Human ES cells’ ability to recombine nonisogenic DNA in a homologous fashion will facilitate the use of the targeting vectors that have already been designed for other human cell lines. Moreover, most mouse ES cell lines were derived from the same mouse strain while many genetically varied human ES cell lines are currently studied. Thus, it may be advantageous to use the same construct in targeting the same gene in different cell lines, even with somewhat less efficient recombination.
In summary, our analysis points to the relevance of human ES cells in complementing murine models for the study of human genetic diseases. It also emphasizes the advanced possibilities of utilizing these cells in investigating developmental diseases.
ACKNOWLEDGMENTS
We thank Dr. Rafael Yanez for sharing with us the construct for the homologous recombination of the HPRT1 gene. We are grateful to Professor Joseph Itskovitz-Eldor at the Rambam Medical Center for his collaboration and for kindly providing us with human ES cells. This study was partially supported by funds from the Herbert Cohn Chair, the Israel Science Foundation, and an NIH grant.
REFERENCES
Thomson JA, Itskovitz-Eldor J, Shapiro SS et al. Embryonic stem cell lines derived from human blastocysts. Science 1998;282:1145–1147.
Itskovitz-Eldor J, Schuldiner M, Karsenti D et al. Differentiation of human embryonic stem cells into embryoid bodies compromising the three embryonic germ layers. Mol Med 2000;6:88–95.
Reubinoff BE, Pera MF, Fong CY et al. Embryonic stem cell lines from human blastocysts: somatic differentiation in vitro. Nat Biotechnol 2000;18:399–404.
Stout JT, Caskey CT. The Lesch-Nyhan syndrome: clinical, molecular and genetic aspects. Trends Genet 1988;4:175–178.
Visser JE, Bar PR, Jinnah HA. Lesch-Nyhan disease and the basal ganglia. Brain Res Brain Res Rev 2000;32:449–475.
Saito Y, Takashima S. Neurotransmitter changes in the pathophysiology of Lesch-Nyhan syndrome. Brain Dev 2000;22(suppl 1):S122–S131.
Bedell MA, Largaespada DA, Jenkins NA et al. Mouse models of human disease. Part II: recent progress and future directions. Genes Dev 1997;11:11–43.
Elsea SH, Lucas RE. The mousetrap: what we can learn when the mouse model does not mimic the human disease. ILAR J 2002;43:66–79.
Zwaka TP, Thomson JA. Homologous recombination in human embryonic stem cells. Nat Biotechnol 2003;21:319–321.
Yanez RJ, Porter AC. Gene targeting is enhanced in human cells overexpressing hRAD51. Gene Ther 1999;6:1282–1290.
Eiges R, Schuldiner M, Drukker M et al. Establishment of human embryonic stem cell transfected clones carrying a marker for undifferentiated cells. Curr Biol 2001;11:514–518.
Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction.Anal Biochem 1987;162:156–159.
Schuldiner M, Eiges R, Eden A et al. Induced neuronal differentiation of human embryonic stem cells. Brain Res 2001;913:201–205.
Sedivy JM, Vogelstein B, Liber HL et al. Gene targeting in human cells without isogenic DNA. Science 1999;283:9.
te Riele H, Maandag ER, Berns A. Highly efficient gene targeting in embryonic stem cells through homologous recombination with isogenic DNA constructs. Proc Natl Acad Sci U S A 1992;89:5128–5132.
van Deursen J, Wieringa B. Targeting of the creatine kinase M gene in embryonic stem cells using isogenic and nonisogenic vectors. Nucleic Acids Res 1992;20:3815–3820.
Zhao Z, Fu YX, Hewett-Emmett D et al. Investigating single nucleotide polymorphism (SNP) density in the human genome and its implications for molecular evolution. Gene 2003;312:207–213.
Sachidanandam R, Weissman D, Schmidt SC et al. A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms. Nature 2001;409:928–933.(Achiya Urbach, Maya Schul)
Key Words. Human ES cells ? Genetic diseases ? Lesch-Nyhan syndrome ? Homologous recombination
Correspondence: Nissim Benvenisty M.D., Ph.D., Hebrew University Department of Genetics, Institute of Life Sciences, Givat-Ram, Jerusalem 91904 Israel. Telephone: 972-2-6586774; Fax: 972-2-6584972; e-mail: nissimb@mail.ls.huji.ac.il
ABSTRACT
Human embryonic stem (ES) cells are pluripotent cells derived from blastocyst-stage embryos. It has been suggested that these cells should play a major role in transplantation medicine and be able to advance our knowledge in human embryology. We propose that these cells should also play a vital role in the creation of models of human disorders. This aspect would be most valuable where animal models failed to faithfully recapitulate the human phenotype. Lesch-Nyhan disease is caused by a mutation in the HPRT1 gene that triggers an overproduction of uric acid, causing gout-like symptoms and urinary stones, in addition to neurological disorders. Due to biochemical differences between humans and rodents, a mouse lacking the HPRT expression will fail to accumulate uric acid. In this research we demonstrate a model for Lesch-Nyhan disease by mutating the HPRT1 gene in human ES cells using homologous recombination. We have verified the mutation in the HPRT1 allele at the DNA and RNA levels. By using selection media, we show that HPRT1 activity is abolished in the mutant cells, and the HPRT1–cells show a higher rate of uric acid accumulation than the wild-type cells. Therefore, these cells recapitulate to some extent the characteristics of Lesch-Nyhan syndrome and can help researchers further investigate this genetic disease and analyze drugs that will prevent the onset of its symptoms. We therefore suggest that human diseases may be modeled using human ES cells.
INTRODUCTION
Human embryonic stem (ES) cells are pluripotent cells derived from the inner cell mass of blastocyst-stage embryos . These cells have two main properties: self-renewal capacity and pluripotent differentiation potential. The pluripotency of the cells has been shown reproducibly in vitro, where human ES cells produce embryonic bodies (EBs) that comprise the three embryonic germ layers . These cells also have the ability for in vivo differentiation as they form differentiated tumors upon injection into immunodeficient severe combined immunodeficiency syndrome/Beige mice . These tumors contain cells from all three germ layers.
These properties of the human ES cells give them a major role in the study of early human development and in medical research because of their potential role in transplantation therapy. However, human ES cells also enable us to generate models for human diseases. In this work we created a model for Lesch-Nyhan disease by inducing a mutation in human ES cells. Lesch-Nyhan syndrome is a severe genetic disease that is caused by the malfunction of the hypoxanthine phosphoribosyltransferase (HPRT1) gene . One main clinical feature of this disease is hyperuricemia that leads to fatal uric acid urinary stones and a gout-like syndrome. In addition, the disease exhibits characteristic neurological symptoms such as mental retardation, spastic cerebral palsy, choreoathetosis, and self-destructive biting of fingers and lips. It is still not clear how the deficiency in HPRT1 activity leads to these neurological symptoms, although it is documented that dopaminergic neurons are damaged in affected patients . The biochemical features leading to overproduction of uric acid are due to an inability to convert hypoxanthine into inosine 5'-monophosphate and the conversion of hypoxanthine into uric acid. In rodents, uric acid does not accumulate because of the activity of the urate oxidase (UOX) enzyme that converts uric acid into allantoin . In addition, no clear neurobehavioral defects have been observed in mouse models that exhibit HPRT1 deficiency. Thus, no good model for Lesch-Nyhan syndrome has been achieved in rodents .
Homologous recombination recently has been described as a method to induce mutation in human ES cells . In order to create a model for Lesch-Nyhan disease in human cells, we have targeted HPRT1 in human ES cells by homologous recombination, using a replacement-type vector that is based on nonisogenic DNA. The mutation in the HPRT1 gene by homologous recombination was confirmed at the DNA and RNA levels and by using selection media to demonstrate an absence in HPRT1 activity. The mutant cells produced significantly more uric acid than wild-type cells do. We believe that these cells will enable further research on Lesch-Nyhan disease, especially in aspects where using mouse models was limited. In addition, our analysis shows that homologous recombination in human ES cells can be achieved with a nonisogenic construct transfected with a cationic reagent.
MATERIALS AND METHODS
Cell Culture
Male human ES cells (H13) were cultured as described by Itskovitz-Eldor et al. using media consisting of 85% KnockoutTMD-MEM (Gibco-Invitrogen Corporation, Carls-bad, CA), 15% KnockoutTMSR—a serum-free formulation (Gibco-Invitrogen Corporation), 1 mM glutamine (Gibco-Invitrogen Corporation, St. Louis), 0.1 mM ?-mercaptoethanol (Sigma), 1% nonessential amino acids stock (Gibco-Invitrogen Corporation), 4 ng/ml basic fibroblast growth factor (bFGF; Gibco-Invitrogen Corporation), and ITS 100X (insulin transferrin selenium), 1:200 dilution (Gibco-Invitrogen Corporation). EBs were formed by transferring 2–4 x 106 ES cells using trypsin/EDTA (0.1%/1 mM) to plastic Petri dishes (Miniplats, Ein-Shemer, Israel) in order to allow their aggregation. The EBs were grown in the same medium as ES cells but without bFGF.
Transfection and Establishment of Mutant Cell Lines
PHPRThyg , a plasmid that contains an insertion of the hygromycin resistance gene into an 8.5-kb fragment of exons II and III of the HPRT1 gene, was introduced into human ES cells using ExGen 500 (Fermentas, Hanover, MD) as described by Eiget et al . More specifically, 106 cells were transfected with 3–4 μg of plasmid liner DNA, centrifuged, and incubated for 30 minutes with the transfection reagent. Twenty-four hours later cells were replated and after 4 days were selected with either 6-thioguanine (6TG, 1 μg/ml; Sigma), hygromycin (100 μg/ml; Roche, Basel, Switzerland), or both.
Selection Media and Cell Proliferation
Selection media were added to wild-type and mutant cells to test for the activity of the purine salvage pathway. Thus, human ES cells were grown with 6TG (1 μg/ml) inhibiting the growth of cells that express HPRT1. Alternatively, the cells were incubated with HAT (1 x 10–4 M hypoxanthine, 4 x 10–7 M aminopterin, 1.6 x 10–5 M thymidine; Sigma), which inhibits the de novo pathway of purine synthesis and thus prevents the growth of cells that do not express HPRT1. One to three days after adding the selection media, we analyzed cell densities by fixating cells with 2.5% glutaraldehyde (Merck, Whitehouse Station, NJ) to the tissue culture plates. The plates were stained with methylene blue (Sigma) and dissolved in 0.1 M boric acid (pH = 8.5). We extracted color with 0.1 N HCl and read emission at 650 nm. Because color density represents cell number, we prepared growth charts based on color emission.
PCR
Total DNA was extracted using EZ-DNA kit (Biological Industries, Beit Haemek, Israel). DNA samples were subjected to polymerase chain reaction (PCR) amplification using the following primers: I: GCAAGTACTCA-GAACAGCTGC; II: TGATGAACCAGGTTATGACC; III: CATCGAAGCTGAAAGCACCAG. The combination of primer I and II should recognize only the wild-type HPRT1 sequence, while the combination of II and III should identity only the mutant sequence (Fig. 1 caption).
Figure 1. Homologous recombination of the HPRT1 gene. (A): Strategy for homologous recombination of HPRT1 and expected polymerase chain reaction (PCR) products from the wild-type (wt) human embryonic stem (ES) cells and from clones mutated in HPRT1 by homologous recombination. I: A specifice primer for the wt cells; II: a primer that is common to the wt and mutant cells; III: a specific primer for the mutant cells. Shown, at the bottom of the scheme (*), the distribution of the single nucleotide polymorphisms along the recombined HPRT1 sequence. (B): PCR analysis of HPRT1 sequences in wt and in mutant ES cells. DNA was isolated from the cells and then subjected to PCR using the primers mentioned above. (C): Reverse transcriptase PCR analysis of wt and mutant cells; cDNA was preformed from mutant and wt RNA and then subjected to PCR using a set of primers that are located in exons I and V of the HPRT1 gene and are expected to identify only the wt HPRT1 mRNA sequence. Note that the HPRT1–clones lack expression of the wt transcript. Primers that amplify the housekeeping gene GAPDH were used as a control.
RNA Isolation and RT-PCR
Total RNA from ES cells was extracted as described by Chomczynski and Sacchi using guanidine thiocyanate followed by phenol-chloroform and isopropanol precipitation. EBs were subjected to homogenization by polytron and then extracted for total RNA by RNeasy? Mini Kit (Qiagen, Valencia, CA). We synthesized cDNA from 1 μg of total RNA using random hexamer (pd6) as primer (Pharmacia Biotech, New Jersey) and M-MLV reverse transcriptase (Gibco-Invitrogen Corporation). We subjected cDNA samples to PCR amplification with the primers shown in Table 1. For each gene, the DNA primers were derived from different exons to ensure that the PCR product represents the specific mRNA transcript and not genomic DNA.
Table 1. Primer sequences for reverse transcription polymerase chain reaction
Quantitation of Uric Acid Production
Mutant and wild-type human ES cells were grown in 4-ml media for 48 hours in the presence or absence of 100 μM hypoxanthine. The medium was concentrated to 1 ml by SpeedVac (Eppendorf, Hamburg, Germany). We assessed the level of uric acid using liquid chromatography.
RESULTS
In order to generate a model for Lesch-Nyhan disease, we decided to mutate the HPRT1 gene in human ES cells. HPRT1 is X-linked, and thus a targeted mutation of the only allele in male human ES cells may lead to biochemical features observed in Lesch-Nyhan patients. We targeted HPRT1 in human ES cells by homologous recombination using a replacement-type vector. This vector contains 8.5 kb of non-isogenic exonic and intronic sequences of the HPRT1 gene, where the hygromycin resistance marker is placed within exon II (Fig. 1A) . According to the National Center for Biotechnology Information (NCBI) database, this sequence contains 12 single nucleotide polymorphisms (SNPs) that are distributed along the entire sequence (Fig. 1A). Homologous recombination of this plasmid leads to HPRT1 inactivation, resulting in resistance to both hygromycin and 6TG, a nucleotide analog that inhibits growth of HPRT1-expressing cells. Out of more than 108 transfected cells, two human ES cell clones that were resistant to both drugs were isolated following ExGen 500–mediated transfection.
We confirmed the presence of the targeted HPRT1 insertion expected in these clones by PCR (Fig. 1B). In addition, we confirmed by reverse transcriptase PCR (RT-PCR) analysis that the mutated cells do not express the wild-type HPRT1 transcript (Fig. 1C). In order to prove that the HPRT1 gene is not activated in the mutant cells, we compared the proliferation rate of wild-type and mutant cells under the selection of 6TG and HAT. HAT medium inhibits the de novo pathway of purine synthesis and prevents growth of cells that lack an active salvage pathway. The mutant cells proliferate in the presence of 6TG and die in HAT medium while wild-type cells behave in an opposite fashion (Fig. 2). These results confirmed our ability to perform insertion mutagenesis by homologous recombination using a cationic reagent.
Figure 2. Analysis of HPRT1 activity using selection media. (A): In order to examine the activity of HPRT1, we used a set of selection media. One medium contains 6-thioguanine (6TG), which prevents growth of all cells in which the purine salvage pathway is active, while the second is the hypoxanthine-aminopterin-thymidine (HAT) medium, which prevents growth of cells that do not have an active purine salvage pathway. Growth curves of wild-type (wt) and mutant cells in the different selection media are shown. Note that the wt cells stop growing in the presence of 6TG, while the mutant cells stop growing in the presence of HAT. (B): Example of mutated cells growing in the different selection media and stained with methylene blue. Abbreviation: O.D., optical density.
Since one of the hallmarks of Lesch-Nyhan syndrome is the accumulation of uric acid, we have analyzed the levels of uric acid in our targeted cells grown in medium with and without the addition of hypoxanthine. Indeed, uric acid significantly accumulates more in mutant cells than in wild-type cells (Fig. 3). Allopurinol, a drug that affects uric acid accumulation in humans, also dramatically reduced uric acid accumulation in the cells (data not shown). Uric acid fails to accumulate in rodent cells mutated in HPRT since these cells express UOX, an enzyme that converts uric acid into allantoin (Fig. 3). Thus, HPRT1-deficient human ES cells recapitulate the biochemical phenotype observed in Lesch-Nyhan cells, while rodent cells failed to do so.
Figure 3. Uric acid production in HPRT1 cells. (A): Schematic pathway of hypoxanthine metabolism into uric acid and inosine 5'-monophosphate (IMP). Note the degradation of uric acid into allantoin occurs only in rodents and not in humans. (B): Uric acid levels (μM) in wild-type and in HPRT1-deficient human embryonic stem cells. Note that the mutant cells accumulate significantly more uric acid with or without the addition of 100 μM hypoxanthine.
The neurological symptoms of Lesch-Nyhan patients have been related to a malfunction in dopaminergic cells . Although it is yet not possible to achieve fully differentiated dopaminergic neurons that express tyrosine hydroxylase from human ES cells , we have shown that like wild-type cells, spontaneously differentiated mutant cells express a key dopaminergic enzyme—dopamine decarboxylase (DDC; Fig. 4).
Figure 4. Analysis of a dopaminergic marker in HPRT–cells. RNA was isolated from 21-day-old embryonic bodies (EBs) and was subjected to reverse transcriptase polymerase chain reaction using the primers that identify either a dopaminergic marker (dopamine decarboxylase ), an embryonic stem cell marker (OCT4), an endoderm differentiation marker (A-feto-protein ), or a housekeeping marker (GAPDH).
DISCUSSION
Models for human disorders are important tools in analyzing the phenotype and potential therapies for genetic diseases. Several approaches were taken in order to achieve such models both in vivo and in vitro. The creation of mutant mice using the homologous recombination technique was vital in generating animal models for many human diseases. However, morphological, developmental, and genetic differences between humans and mice posed significant limitations for the use of these mutant animals as models for human disorders. In many cases, the mouse phenotype largely differs from those observed in humans . In addition to in vivo animal models, researchers may study in vitro human systems to model human disorders. One such approach is the generation of primary cultures from patients’ cells. This methodology is limited due to the specific range of tissues from which cells can be achieved and the short life span of these primary cultures. In addition, these systems rely on mutations that occurred naturally in humans in a specific genetic background. A way of overcoming this limitation is to induce mutations in human cell lines by choice. Thus, homologous recombination has been exercised on various cell lines mutating mostly genes involved in oncogenesis and senescence . However, the mutations were induced in transformed cells, and in each case, the phenotype was studied in already differentiated, committed cell lines.
Human ES cells are nontransformed cells with the capacities of self-renewal and pluripotency, allowing them to differentiate into a wide range of cell types. So far the unique features of these cells were mainly used in the search for new sources of cell transplantation. However, via genetic manipulation, human ES cells can also be used to model human diseases. This is especially important where the animal in vivo models have failed to fully recapitulate the phenotype in the human disease.
A good example for such a case is Lesch-Nyhan syndrome. This syndrome is caused by accumulation of uric acid due to a mutation in the HPRT1 enzyme . Lesch-Nyhan patients accumulate uric acid and thus suffer from renal failure and a gout-like syndrome, in addition to neurobehavioral symptoms related to a malfunction of the dopaminergic neurons. Uric acid fails to accumulate in mice lacking HPRT activity since they express UOX, the enzyme that converts uric acid into allantoin (Fig. 3) . Thus, it is difficult, if not impossible, to create a faithful model for Lesch-Nyhan syndrome in mice. In our study, we have shown that directed mutagenesis in the HPRT1 gene in human ES cells results in high accumulation of uric acid compared to wild-type cells. Thus, we were able to generate a model for the biochemical phenotype of Lesch-Nyhan syndrome in a culture system. In our mutant cells, the drug allopurinol leads to a decrease in uric acid levels. Therefore, this system may be used, in the future, in the search for new medications for Lesch-Nyhan patients.
Since a significant phenotype in Lesch-Nyhan patients is related to dopaminergic cells, a major challenge is to find a way to induce complete dopaminergic differentiation in human ES cells. Achievement of this goal will enable us to examine biochemical and electrophysiological features in wild-type and mutant dopaminergic cells. Establishment of an in vitro system that mimics the development of undifferentiated HPRT1–cells to dopaminergic neurons may give an opportunity, which does not exist in any other models, to study the developmental processes that lead to the malfunction of the dopaminergic neurons in Lesch-Nyhan syndrome. This system should also allow a better understanding of these patients’neurological symptoms. Although at present conditions in which to differentiate human ES cells to dopaminergic cells are still unknown , the fact that the genetic manipulation of the cells did not prevent their partial differentiation into DDC-expressing cells encourages additional work with these mutant cells in order to further study the neurological phenotype of Lesch-Nyhan syndrome.
In our analysis, we have recombined sequences that were derived from separate human cell lines. A number of studies in mouse ES cells have shown that the use of nonisogenic sequences decreases the efficiency of homologous recombination by up to 20-fold . However, it has been suggested that the efficiency of gene targeting in human cells is the same for isogenic and nonisogenic sequences . In this study, the efficiency of nonhomologous recombination as determined by resistance to an antibiotic selection marker is about 10–6 transfected cells, while the efficiency of homologous recombination is about 10–8 transfected cells. Hence, the ratio of homologous recombination versus nonhomologous recombination in our system is estimated to be about 1%, which is only somewhat lower than that of 2% as recently reported for gene targeting with isogenic sequences by electroporation in human ES cells . Thus, our results support the notion presented by Sedivy et al. that homologous recombination in humans is efficient even with non-isogenic DNA. One possible explanation for the differences between humans and mice, in relation to the efficiency of nonisogenic homologous recombination, is that the sequence divergence between two human cell lines is much smaller than the difference between two different mouse strains. According to the NCBI database, there are 12 SNPs in the 8.5-kb sequences of HPRT1 used in our study for targeted recombination (1 SNP per 0.7 kb). This figure reflects the normal distribution of SNPs in the human genome among nonisogenic sequences . However, these differences are much smaller than those that appear in the sequences of different mice strains . Human ES cells’ ability to recombine nonisogenic DNA in a homologous fashion will facilitate the use of the targeting vectors that have already been designed for other human cell lines. Moreover, most mouse ES cell lines were derived from the same mouse strain while many genetically varied human ES cell lines are currently studied. Thus, it may be advantageous to use the same construct in targeting the same gene in different cell lines, even with somewhat less efficient recombination.
In summary, our analysis points to the relevance of human ES cells in complementing murine models for the study of human genetic diseases. It also emphasizes the advanced possibilities of utilizing these cells in investigating developmental diseases.
ACKNOWLEDGMENTS
We thank Dr. Rafael Yanez for sharing with us the construct for the homologous recombination of the HPRT1 gene. We are grateful to Professor Joseph Itskovitz-Eldor at the Rambam Medical Center for his collaboration and for kindly providing us with human ES cells. This study was partially supported by funds from the Herbert Cohn Chair, the Israel Science Foundation, and an NIH grant.
REFERENCES
Thomson JA, Itskovitz-Eldor J, Shapiro SS et al. Embryonic stem cell lines derived from human blastocysts. Science 1998;282:1145–1147.
Itskovitz-Eldor J, Schuldiner M, Karsenti D et al. Differentiation of human embryonic stem cells into embryoid bodies compromising the three embryonic germ layers. Mol Med 2000;6:88–95.
Reubinoff BE, Pera MF, Fong CY et al. Embryonic stem cell lines from human blastocysts: somatic differentiation in vitro. Nat Biotechnol 2000;18:399–404.
Stout JT, Caskey CT. The Lesch-Nyhan syndrome: clinical, molecular and genetic aspects. Trends Genet 1988;4:175–178.
Visser JE, Bar PR, Jinnah HA. Lesch-Nyhan disease and the basal ganglia. Brain Res Brain Res Rev 2000;32:449–475.
Saito Y, Takashima S. Neurotransmitter changes in the pathophysiology of Lesch-Nyhan syndrome. Brain Dev 2000;22(suppl 1):S122–S131.
Bedell MA, Largaespada DA, Jenkins NA et al. Mouse models of human disease. Part II: recent progress and future directions. Genes Dev 1997;11:11–43.
Elsea SH, Lucas RE. The mousetrap: what we can learn when the mouse model does not mimic the human disease. ILAR J 2002;43:66–79.
Zwaka TP, Thomson JA. Homologous recombination in human embryonic stem cells. Nat Biotechnol 2003;21:319–321.
Yanez RJ, Porter AC. Gene targeting is enhanced in human cells overexpressing hRAD51. Gene Ther 1999;6:1282–1290.
Eiges R, Schuldiner M, Drukker M et al. Establishment of human embryonic stem cell transfected clones carrying a marker for undifferentiated cells. Curr Biol 2001;11:514–518.
Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction.Anal Biochem 1987;162:156–159.
Schuldiner M, Eiges R, Eden A et al. Induced neuronal differentiation of human embryonic stem cells. Brain Res 2001;913:201–205.
Sedivy JM, Vogelstein B, Liber HL et al. Gene targeting in human cells without isogenic DNA. Science 1999;283:9.
te Riele H, Maandag ER, Berns A. Highly efficient gene targeting in embryonic stem cells through homologous recombination with isogenic DNA constructs. Proc Natl Acad Sci U S A 1992;89:5128–5132.
van Deursen J, Wieringa B. Targeting of the creatine kinase M gene in embryonic stem cells using isogenic and nonisogenic vectors. Nucleic Acids Res 1992;20:3815–3820.
Zhao Z, Fu YX, Hewett-Emmett D et al. Investigating single nucleotide polymorphism (SNP) density in the human genome and its implications for molecular evolution. Gene 2003;312:207–213.
Sachidanandam R, Weissman D, Schmidt SC et al. A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms. Nature 2001;409:928–933.(Achiya Urbach, Maya Schul)