Single-cell RNA sequencing of neural stem cells derived from human trisomic iPSCs reveals the abnormalities during neural differentiation of Down syndrome.

Introduction: Down syndrome (DS) is the most common genetic condition that causes intellectual disability in humans. The molecular mechanisms behind the DS phenotype remain unclear. Therefore, in this study, we present new findings on its molecular mechanisms through single-cell RNA sequencing. Methods: Induced pluripotent stem cells (iPSCs) from the patients with DS and the normal control (NC) patients were differentiated into iPSCs-derived neural stem cells (NSCs). Single-cell RNA sequencing was performed to achieve a comprehensive single-cell level differentiation roadmap for DS-iPSCs. Biological experiments were also performed to validate the findings. Results and Discussion: The results demonstrated that iPSCs can differentiate into NSCs in both DS and NC samples. Furthermore, 19,422 cells were obtained from iPSC samples (8,500 cells for DS and 10,922 cells for the NC) and 16,506 cells from NSC samples (7,182 cells for DS and 9,324 cells for the NC), which had differentiated from the iPSCs. A cluster of DS-iPSCs, named DS-iPSCs-not differentiated (DSi-PSCs-ND), which had abnormal expression patterns compared with NC-iPSCs, were demonstrated to be unable to differentiate into DS-NSCs. Further analysis of the differentially expressed genes revealed that inhibitor of differentiation family (ID family) members, which exhibited abnormal expression patterns throughout the differentiation process from DS-iPSCs to DS-NSCs, may potentially have contributed to the neural differentiation of DS-iPSCs. Moreover, abnormal differentiation fate was observed in DS-NSCs, which resulted in the increased differentiation of glial cells, such as astrocytes, but decreased differentiation into neuronal cells. Furthermore, functional analysis demonstrated that DS-NSCs and DS-NPCs had disorders in axon and visual system development. The present study provided a new insight into the pathogenesis of DS.

Missense mutation of c.635 T > C in CAPN3 impairs muscle injury repair in a Limb-Girdel Muscular Dystropy Model.

Limb-girdle muscular dystrophy recessive 1 (LGMDR1), previously known as LGMD2A, is a specific LGMD caused by a gene mutation encoding the calcium-dependent neutral cysteine protease calpain-3 (CAPN3). In our study, the compound heterozygosity with two missense variants c.635 T > C (p.Leu212Pro) and c.2120A > G (p.Asp707Gly) was identified in patients with LGMDR1. However, the pathogenicity of c.635 T > C has not been investigated. To evaluate the effects of this novel likely pathogenic variant to the motor system, the mouse model with c.635 T > C variant was prepared by CRISPR/Cas9 gene editing technique. The pathological results revealed that a limited number of inflammatory cells infiltrated the endomyocytes of certain c.635 T > C homozygous mice at 10 months of age. Compared with wild-type mice, motor function was not significantly impaired in Capn3 c. 635 T > C homozygous mice. Western blot and immunofluorescence assays further indicated that the expression levels of the Capn3 protein in muscle tissues of homozygous mice were similar to those of wild-type mice. However, the arrangement and ultrastructural alterations of the mitochondria in the muscular tissues of homozygous mice were confirmed by electron microscopy. Subsequently, muscle regeneration of LGMDR1 was simulated using cardiotoxin (CTX) to induce muscle necrosis and regeneration to trigger the injury modification process. The repair of the homozygous mice was significantly worse than that of the control mice at day 15 and day 21 following treatment, the c.635 T > C variant of Capn3 exhibited a significant effect on muscle regeneration of homozygous mice and induced mitochondrial damage. RNA-sequencing results demonstrated that the expression levels of the mitochondrial-related functional genes were significantly downregulated in the mutant mice. Taken together, the results of the present study strongly suggested that the LGMDR1 mouse model with a novel c.635 T > C variant in the Capn3 gene was significantly dysfunctional in muscle injury repair via impairment of the mitochondrial function.

Genome-wide hypermethylation is closely associated with abnormal expression of genes involved in neural development in induced pluripotent stem cells derived from a Down syndrome mouse model.

Mental retardation is the main clinical manifestation of Down syndrome (DS), and neural abnormalities occur during the early embryonic period and continue throughout life. Tc1, a model mouse for DS, carries the majority part of the human chromosome 21 and has multiple neuropathy phenotypes similar to patients with DS. To explore the mechanism of early neural abnormalities of Tc1 mouse, induced pluripotent stem (iPS) cells from Tc1 mice were obtained, and genome-wide gene expression and methylation analysis were performed for Tc1 and wild-type iPS cells. Our results showed hypermethylation profiles for Tc1 iPS cells, and the abnormal genes were shown to be related to neurodevelopment and distributed on multiple chromosomes. In addition, important genes involved in neurogenesis and neurodevelopment were shown to be downregulated in Tc1 iPS cells. In short, our study indicated that genome-wide hypermethylation leads to the disordered expression of genes associated with neurodevelopment in Tc1 mice during early development. Overall, our work provided a useful reference for the study of the molecular mechanism of nervous system abnormalities in DS.

Dysfunctions of mitochondria in close association with strong perturbation of long noncoding RNAs expression in down syndrome.

Trisomy 21 is the most common chromosomal disorder and underlies Down syndrome. Epigenetics, such as DNA methylation and post-translational histone modifications, plays a vital role in Down syndrome. However, the functions of epigenetics-related long noncoding RNAs (lncRNAs), found to have an impact on neural diseases such as Alzheimer's disease, remain unknown in Down syndrome. In this study, we analyzed the RNA sequencing data from Down syndrome-induced pluripotent stem cells (iPSCs) and normal iPSCs. A large number of lncRNAs were identified differentially expressed in Down syndrome-iPSCs. Notably, stronger perturbation was shown in the expression of lncRNAs compared to protein coding genes (Kolmogorov-Smirnov test, P<0.05), suggesting that lncRNAs play more important roles in Down syndrome. Through gene set enrichment analysis and bi-clustering, we also found that most of the differential expressed lncRNAs were closely associated with mitochondrial functions (e.g. mitochondrion organization, P=3.21×10-17; mitochondrial ATP synthesis coupled electron transport, P=1.73×10-19 and mitochondrial membrane organization, P=4.04×10-8). PCR-array and qRT-PCR results revealed that almost all genes related to mitochondria were down-regulated in Down syndrome-iPSCs, implying that mitochondria were dysfunctional in Down syndrome (e.g. ATP5B, Fold Change=-8.2317; COX6A1, Fold Change=-12.7788 and SLC25A17, Fold Change=-22.1296). All in all, our study indicated that a stronger perturbation of lncRNAs expression may lead to the dysfunction of mitochondria in Down syndrome.

Detection of fetal epigenetic biomarkers through genome-wide DNA methylation study for non-invasive prenatal diagnosis.

The discovery of cell-free DNA fetal (cff DNA) in maternal plasma during pregnancy provides a novel perspective for the development of non­invasive prenatal diagnosis (NIPD). Against the background of maternal DNA, the use of the relatively low concentration of cff DNA is limited in NIPD. Therefore, in order to overcome the complication of the background of maternal DNA and expand the scope of cff DNA application in clinical practice, it is necessary to identify novel universal fetal­specific DNA markers. The GeneChip Human Promoter 1.0R Array set was used in the present study to analyze the methylation status of 12 placental tissue and maternal peripheral blood whole­genome DNA samples. In total, 5 fetus differential hypermethylation regions and 6 fetus differential hypomethylation regions were identified. In order to verify the 11 selected methylation regions and detect the differential CpG sites in these regions, a bisulfate direct sequencing strategy was used. In total, 87 fetal differential methylation CpG sites were identified from 123 CpG sites. The detection of fetal differential methylation DNA regions and CpG sites may be instrumental in the development of efficient NIPD and in the expansion of its application in other disorders.

Identification and functional analysis of long non-coding RNAs in human and mouse early embryos based on single-cell transcriptome data.

Epigenetics regulations have an important role in fertilization and proper embryonic development, and several human diseases are associated with epigenetic modification disorders, such as Rett syndrome, Beckwith-Wiedemann syndrome and Angelman syndrome. However, the dynamics and functions of long non-coding RNAs (lncRNAs), one type of epigenetic regulators, in human pre-implantation development have not yet been demonstrated. In this study, a comprehensive analysis of human and mouse early-stage embryonic lncRNAs was performed based on public single-cell RNA sequencing data. Expression profile analysis revealed that lncRNAs are expressed in a developmental stage-specific manner during human early-stage embryonic development, whereas a more temporal-specific expression pattern was identified in mouse embryos. Weighted gene co-expression network analysis suggested that lncRNAs involved in human early-stage embryonic development are associated with several important functions and processes, such as oocyte maturation, zygotic genome activation and mitochondrial functions. We also found that the network of lncRNAs involved in zygotic genome activation was highly preservative between human and mouse embryos, whereas in other stages no strong correlation between human and mouse embryo was observed. This study provides insight into the molecular mechanism underlying lncRNA involvement in human pre-implantation embryonic development.

Long non-coding RNA LINC01296 is a potential prognostic biomarker in patients with colorectal cancer.

Colorectal cancer (CRC), one of the most malignant cancers, is currently the fourth leading cause of cancer deaths worldwide. Recent studies indicated that long non-coding RNAs (lncRNAs) could be robust molecular prognostic biomarkers that can refine the conventional tumor-node-metastasis staging system to predict the outcomes of CRC patients. In this study, the lncRNA expression profiles were analyzed in five datasets (GSE24549, GSE24550, GSE35834, GSE50421, and GSE31737) by probe set reannotation and an lncRNA classification pipeline. Twenty-five lncRNAs were differentially expressed between CRC tissue and tumor-adjacent normal tissue samples. In these 25 lncRNAs, patients with higher expression of LINC01296, LINC00152, and FIRRE showed significantly better overall survival than those with lower expression (P < 0.05), suggesting that these lncRNAs might be associated with prognosis. Multivariate analysis indicated that LINC01296 overexpression was an independent predictor for patients' prognosis in the test datasets (GSE24549, GSE24550) (P = 0.001) and an independent validation series (GSE39582) (P = 0.027). Our results suggest that LINC01296 could be a novel prognosis biomarker for the diagnosis of CRC.

Alteration of gene expression profiling including GPR174 and GNG2 is associated with vasovagal syncope.

Vasovagal syncope (VVS) causes accidental harm for susceptible patients. However, pathophysiology of this disorder remains largely unknown. In an effort to understanding of molecular mechanism for VVS, genome-wide gene expression profiling analyses were performed on VVS patients at syncope state. A total of 66 Type 1 VVS child patients and the same number healthy controls were enrolled in this study. Peripheral blood RNAs were isolated from all subjects, of which 10 RNA samples were randomly selected from each groups for gene expression profile analysis using Gene ST 1.0 arrays (Affymetrix). The results revealed that 103 genes were differently expressed between the patients and controls. Significantly, two G-proteins related genes, GPR174 and GNG2 that have not been related to VVS were among the differently expressed genes. The microarray results were confirmed by qRT-PCR in all the tested individuals. Ingenuity pathway analysis and gene ontology annotation study showed that the differently expressed genes are associated with stress response and apoptosis, suggesting that the alteration of some gene expression including G-proteins related genes is associated with VVS. This study provides new insight into the molecular mechanism of VVS and would be helpful to further identify new molecular biomarkers for the disease.

Pyrosequencing is an accurate and reliable method for the analysis of heteroplasmy of the A3243G mutation in patients with mitochondrial diabetes.

Human mitochondrial DNA is a circular DNA molecule that encodes some of the proteins required for oxidative phosphorylation. Different mitochondrial DNA genotypes may coexist within a single cell, a condition known as heteroplasmy. An A-to-G transition at position 3243 of mitochondrial DNA (A3243G) can result in maternally inherited diabetes and deafness (mitochondrial diabetes). However, the commonly used methods of PCR restriction fragment length polymorphism and Sanger sequencing are neither sensitive nor reliable enough to detect this low level of heteroplasmy. Here, we developed a quantitative method based on pyrosequencing to analyze the heteroplasmy of the A3243G mutation in leukocyte DNA obtained from 83 persons of 15 unrelated pedigrees with mitochondrial diabetes. The accuracy and reliability of this method were also measured by comparing the results with those from high-resolution melting analysis, Sanger sequencing, and PCR restriction fragment length polymorphism with artificial heteroplasmy standard samples. The results showed that the accuracy of pyrosequencing was much higher than that of the other methods, and the limitation of heteroplasmy detection with this method reached 2%, based on our artificial control studies. An inverse correlation was found between the level of heteroplasmy and the age of the onset in our patients. This result suggested that the heteroplasmy of the A3243G mutation could become a significant prediction index for the onset of mitochondrial diabetes.

Rapid screening for chromosomal aneuploidies using array-MLPA.

BACKGROUND: Chromosome abnormalities, especially trisomy of chromosome 21, 13, or 18 as well as sex chromosome aneuploidy, are a well-established cause of pregnancy loss. Cultured cell karyotype analysis and FISH have been considered reliable detectors of fetal abnormality. However, results are usually not available for 3-4 days or more. Multiplex ligation-dependent probe amplification (MLPA) has emerged as an alternative rapid technique for detection of chromosome aneuploidies. However, conventional MLPA does not allow for relative quantification of more than 50 different target sequences in one reaction and does not detect mosaic trisomy. A multiplexed MLPA with more sensitive detection would be useful for fetal genetic screening. METHODS: We developed a method of array-based MLPA to rapidly screen for common aneuploidies. We designed 116 universal tag-probes covering chromosomes 13, 18, 21, X, and Y, and 8 control autosomal genes. We performed MLPA and hybridized the products on a 4-well flow-through microarray system. We determined chromosome copy numbers by analyzing the relative signals of the chromosome-specific probes. RESULTS: In a blind study of 161 peripheral blood and 12 amniotic fluid samples previously karyotyped, 169 of 173 (97.7%) including all the amniotic fluid samples were correctly identified by array-MLPA. Furthermore, we detected two chromosome X monosomy mosaic cases in which the mosaism rates estimated by array-MLPA were basically consistent with the results from karyotyping. Additionally, we identified five Y chromosome abnormalities in which G-banding could not distinguish their origins for four of the five cases. CONCLUSIONS: Our study demonstrates the successful application and strong potential of array-MLPA in clinical diagnosis and prenatal testing for rapid and sensitive chromosomal aneuploidy screening. Furthermore, we have developed a simple and rapid procedure for screening copy numbers on chromosomes 13, 18, 21, X, and Y using array-MLPA.

Application of high-resolution melting for genotyping bovine mitochondrial DNA.

Recent studies have demonstrated that mitochondrial DNA (mtDNA) haplotype has a significant impact on the efficiency of bovine somatic cell nuclear transfer. Conventional methods for detecting mtDNA variations and haplotypes, such as restriction fragment length polymorphism (RFLP), temporal temperature gradient gel electrophoresis, dHPLC and sequencing, are labor intensive or expensive and have low sensitivity. High-resolution melting (HRM) analysis is a new technique for mutation detection and has the advantages of speed, cost, and accuracy. Here, we describe the genotyping of bovine mtDNA using HRM analysis. DNA samples containing mtDNA were extracted from 75 Holstein cows and subjected to rapid-cycle (<20 min) PCR of small amplicons (<120 bp) using specific primer sets. Capillaries containing the PCR products were then subjected to HRM analysis; data were acquired in 2 min and analyzed using the instrument's software. Five common bovine mtDNA single nucleotide polymorphisms were identified: 9602 G>A, 169 A>G, 166A>G with 173A>G, and 363C>G. These results agree with both sequencing and RFLP analysis. In addition, a very small amount of heteroplasmic variants (<5%) was sufficiently to be distinguished by HRM analysis that would be very useful to differentiate heteroplasmy vs. homoplasmy. HRM analysis thus provides a new approach to genotyping bovine mtDNA sequence variations and has many advantages over other methods, including speed of analysis, cost, and accuracy. We believe this will be a valuable technique for determining the efficiency of nuclear transfer in cloned embryos and for studying maternal effects on nuclear-cytoplasm interactions.

Rapid and reliable detection of glucose-6-phosphate dehydrogenase (G6PD) gene mutations in Han Chinese using high-resolution melting analysis.

Glucose-6-phosphate dehydrogenase (G6PD) deficiency, an X-linked inherited disease, is one of the most common enzymopathies and affects over 400 million people worldwide. In China at least 21 distinct point mutations have been identified so far. In this study high-resolution melting (HRM) analysis was used to screen for G6PD mutations in 260 unrelated Han Chinese individuals, and the rapidity and reliability of this method was investigated. The mutants were readily differentiated by using HRM analysis, which produced distinct melting curves for each tested mutation. Interestingly, G1388A and G1376T, the two most common variants accounting for 50% to 60% of G6PD deficiency mutations in the Chinese population, could be differentiated in a single reaction. Further, two G6PD mutations not previously reported in the Chinese population were identified in this study. One of these mutations, designated "G6PD Jiangxi G1340T," involved a G1340T substitution in exon 11, predicting a Gly447Val change in the protein. The other mutation involved a C406T substitution in exon 5. The frequencies of the common polymorphism site C1311T/IVS (intervening sequence) XI t93c between patients with G6PD and healthy volunteers were not significantly different. Thus, HRM analysis will be a useful alternative for screening G6PD mutations.

A phiC31 integrase-mediated integration hotspot in favor of transgene expression exists in the bovine genome.

phiC31 integrase, a site-specific recombinase, can effectively mediate foreign genes bearing an attB sequence integrated into pseudo attP sites. We have previously identified two pseudo attP sites, BpsF1 and BpsM1 from the bovine genome. In this study, two new pseudo attP sites, BF4 and BF10, were discovered using half-nested inverse PCR from cow fibroblasts. The genomic locations of these two pseudo attP sites were identified by direct sequencing and a BLAST search, and it was confirmed that they reside at positions 4q31 and 10q35 by fluorescence in situ hybridization analysis. Subsequently, the distinct integration frequencies of the four pseudo attP sites were examined. The BF4 site was identified as a hotspot where site-specific integration occurred in most of the cell clones examined, accounting for 74% (42/57) of the integration; much more than the integration frequency for BF10 (7%; 4/57), BpsF1 (7%; 4/57) and BpsM1 (0/57). Interestingly, similar to other hotspots identified in the human and mouse genomes, in which transgenes integrated at hotspots result in high expression, the GFP gene integrated at hotspot BF4 was expressed at high levels in cow fibroblasts, as confirmed by fluorescence microscopy and FACS analysis. Furthermore, ELISA showed that the expression level of the GFP gene integrated at the BF4 site averaged approximately 328 microg x mg(-1), which is more than twofold higher than that integrated at the BF10 site. This study suggests that somatic cells carrying a desired gene integrated at the BF4 site can be used as nuclear donors to generate valuable transgenic animals by nuclear transfer.

[Preliminary application of MLPA-based array in detecting Y chromosome abnormalities].

To explore the feasibility and accuracy of MLPA-based array (Array-MLPA) in detecting sex chromosome abnormalities, MLPA probes were designed to target against three gene loci, TSPY (p11.2), PRY (q11), and RBMY (q11.2) in human Y chromosome. Array-MLPA approach was applied to test abnormalities of Y chromosome in 15 patient samples with known karyotypes. The data were compared with karyotyping and PCR analyses. The results showed that the copy number of each site detected by Array-MLPA was basically consistent with karyotyping analysis. Moreover, small deletions of chromosomes that were not found by routine karyotyping analysis were identified by the approach described, which fully agreed with PCR analysis, indicating that Array-MLPA was able to detect small abnormalities of chromosomes that cannot be found by karyotyping analysis. Compared to the routine karyotyping method, Array-MLPA has the advantages of high efficiency and reliability in chromosomal analysis, which has great potential in clinical application of diagnosis of chromosome abnormalities.

[Computational and experimental identification of novel microRNAs in goat.].

microRNA (miRNA) is a class of non-coding small RNA molecules with roughly 22 nucleotides in length, regulating gene expression on post-transcriptional level and playing an important role in cell proliferation, differentiation and apoptosis process. Based on the conservation of miRNAs sequence, we compared the known miRNAs among five mammals, i.e., human, mouse, cattle, pig and dog with the sequence of sheep genome that is highly homologous to goat genome, published on the NCBI, and 11 candidate miRNAs were eventually obtained. RT-PCR analysis showed the expression of the 11 miRNAs in brain and 5 in liver, indicating that they might be novel miRNAs. The methodology provides an alternative approach to the exploration of new miRNAs in goat.

Array-MLPA: comprehensive detection of deletions and duplications and its application to DMD patients.

Multiplex ligation-dependent probe amplification (MLPA) is widely used to screen genes of interest for deletions and duplications. Since MLPA is usually based on size-separation of the amplification products, the maximum number of target sequences that can be screened in parallel is usually limited to approximately 40. We report the design of a robust array-based MLPA format that uses amplification products of essentially uniform size (100-120 bp) and distinguishes between them by virtue of incorporated tag sequences. We were thus able to increase probe complexity to 124, with very uniform product yields and signals that have a low coefficient of variance. The assay designed was used to screen the largest set studied so far (249 patients) of unrelated Duchenne muscular dystrophy (DMD) cases from the Chinese population. In a blind study we correctly assigned 98% of the genotypes and detected rearrangements in 181 cases (73%); i.e., 163 deletions (65%), 13 duplications (5%), and five complex rearrangements (2%). Although this value is significantly higher for Chinese patients than previously reported, it is similar to that found for other populations. The location of the rearrangements (76% in the major deletion hotspot) is also in agreement with other findings. The 96-well flow-through microarray system used in this research provides high-throughput and speed; hybridization can be completed in 5 to 30 minutes. Since array processing and data analysis are fully automated, array-MLPA should be easy to implement in a standard diagnostic laboratory. The universal array can be used to analyze any tag-modified MLPA probe set.

Effect of oocyte mitochondrial DNA haplotype on bovine somatic cell nuclear transfer efficiency.

The development capability of reconstructed bovine embryos via ovum pick-up (OPU)-somatic cell nuclear transfer (SCNT) technique has been influenced by the maternal lineage of oocyte cytoplasm, but the underlying mechanism remains unclear. Since mitochondria are the richest maternal-inherited organelle, in this study, we intended to clarify the effect of mtDNA haplotypes on cloning efficiency. By PCR-RFLP method, we identified mtDNA haplotypes A and B, differing in six restriction sites. Reconstructed embryos with haplotype A cytoplast achieved better fusion and blastocyst formation rate (64.6% and 39.4%), as compared with haplotype B (53.6% and 26.3%; P < 0.05). To further evaluate the role of mitochondria, the quantity of mtDNA, ATP content, and mRNA level of mtDNA-encoded COXI, COXIII in both oocytes were measured. Our data indicated that mtDNA copy number in haplotype A oocyte was significantly higher than that in haplotype B oocyte, both at the GV (10(5.03 +/- 0.69) vs. 10(4.81 +/- 0.86) copies/oocyte) and MII stages (10(5.31 +/- 0.71) vs. 10(5.13 +/- 0.63) copies/oocyte; logarithmically transformed values; P < 0.05). ATP content in type A oocyte was also greater at the GV (1.67 +/- 0.09 vs. 1.27 +/- 0.1 pmol) and MII stages (5.18 +/- 0.07 vs. 2.68 +/- 0.03 pmol; P < 0.05). Similarly, the mRNA expression level of mtDNA-encoded COXI and COXIII in haplotype A oocyte was significantly higher comparing to haplotype B oocyte (3.3 +/- 2.0 x 10(3) vs. 0.68 +/- 0.45 x 10(3); 24.9 +/- 10.5 x 10(3) vs. 9.4 +/- 3.3 x 10(3), respectively; P < 0.05). The data suggest that mitochondrial structure, quantity, and function may significantly affect the developmental competence of reconstructed embryos.

Improved efficiency of bovine cloning by autologous somatic cell nuclear transfer.

Somatic cell nuclear transfer (SCNT) has been used for the cloning of various mammals. However, the rates of successful, healthy birth are generally poor. To improve cloning efficiency, we report the utilization of an 'autologous SCNT' cloning technique in which the somatic nucleus of a female bovine donor is transferred to its own enucleated oocyte recovered by ovum pick up, in contrast to the routine 'allogeneic SCNT' procedure using oocytes from unrelated females. Our results showed that embryos derived from autologous SCNThave significantly higher developmental competence than those derived from allogeneic SCNT, especiallyat the eight-cell (60 vs 44%), morula (45 vs 36%), and blastocyst (38 vs 23%) stages. The pregnancy and birth rates were also higher for the autologous (39 and 23%), compared to the allogeneic (22 and 6%) SCNT groups. Genome-wide histone3-lysine9 methylation profiles reveal that autologous SCNTembryos have less epigenetic defects than the allogeneic SCNTembryos. This study indicates that autologous SCNT can improve the efficiency of bovine cloning with less reprogramming deficiency.

Transgenic mice can express mutant human coagulation factor IX with higher level of clotting activity.

To improve the available values of transgenic animals, we produced a mutant human coagulation factor IX minigene (including cDNA and intron I) with arginine at 338 changed to alanine (R338A-hFIX) by using a direct mutation technique. The R338A-hFIX minigene was then cloned into a plasmid carrying the goat beta-casein promoter to get a mammary gland-specific expression vector. The clotting activity in the supernatant of the transfected HC-11 cells increased to approximately three times more than that of wild-type hFIX. Nine transgenic mice (three females and six males) were produced, and the copy number of the foreign gene was very different, ranging from 1 to 43 in different lines. ELISA, Western blot, and clotting assay experiments showed that the transgenic mice could express R338A-hFIX, showing higher average levels of clotting activity than wild-type hFIX in the milk (103.76% vs. 49.95%). The highest concentration and clotting activity of hFIX reached 26 mug/mL and 1287% in one founder (F(0)-7), which was over 10 times higher than that in human plasma. Furthermore, RT-PCR, APTT assay, and histological analysis indicated that hFIX was expressed specifically in the mammary gland without affecting the intrinsic coagulation pathway and physiologic performance of the local tissue.