The evolving privacy and security concerns for genomic data analysis and sharing as observed from the iDASH competition

Concerns regarding inappropriate leakage of sensitive personal information as well as unauthorized data use are increasing with the growth of genomic data repositories. Therefore, privacy and security of genomic data have become increasingly important and need to be studied. With many proposed protection techniques, their applicability in support of biomedical research should be well understood. For this purpose, we have organized a community effort in the past 8 years through the integrating data for analysis, anonymization and sharing consortium to address this practical challenge. In this article, we summarize our experience from these competitions, report lessons learned from the events in 2020/2021 as examples, and discuss potential future research directions in this emerging field.

Haplotype-based membership inference from summary genomic data

Motivation

The availability of human genomic data, together with the enhanced capacity to process them, is leading to transformative technological advances in biomedical science and engineering. However, the public dissemination of such data has been difficult due to privacy concerns. Specifically, it has been shown that the presence of a human subject in a case group can be inferred from the shared summary statistics of the group, e.g. the allele frequencies, or even the presence/absence of genetic variants (e.g. shared by the Beacon project) in the group. These methods rely on the availability of the target’s genome, i.e. the DNA profile of a target human subject, and thus are often referred to as the membership inference method.

Results

In this article, we demonstrate the haplotypes, i.e. the sequence of single nucleotide variations (SNVs) showing strong genetic linkages in human genome databases, may be inferred from the summary of genomic data without using a target’s genome. Furthermore, novel haplotypes that did not appear in the database may be reconstructed solely from the allele frequencies from genomic datasets. These reconstructed haplotypes can be used for a haplotype-based membership inference algorithm to identify target subjects in a case group with greater power than existing methods based on SNVs.

Availability and implementation

The implementation of the membership inference algorithms is available at https://github.com/diybu/Haplotype-based-membership-inferences.

Real-time Protection of Genomic Data Sharing in Beacon Services

The acquisition of human genomic sequences is of increasing convenience and reduced expense. The sharing of these data is critical for biomedical researchers to study genomic loci or variants that are potentially associated with human diseases¹. However, sharing genomic data broadly is impeded by privacy concerns. The statistical inference techniques for the re-identification of genomic data donors have been extensively investigated in the literature^2-5. The Beacon services project is recently brought into view, aiming to test the willingness of data holders to share genomic data in a simple technical context: a query to ask a specified nucleotide at a given position within a chromosome⁶, also suffering from being compromised^7,8. In this paper, we introduce a real-time mitigation method to protect Beacon services from re-identification attacks⁷, and show that it performs favorably in comparison with previous approaches on mitigation efficiency, i.e., with lower re-identification risks and higher utility of Beacon database.

Insights into the epidemic characteristics and evolutionary history of the novel porcine circovirus type 3 in southern China

Porcine circovirus type 3 (PCV3) is a newly identified circovirus from swine in the USA, China and Poland. This novel circovirus has been associated with porcine dermatitis and nephropathy syndrome (PDNS), reproductive failure and multisystemic inflammation; moreover, PCV3 poses a potential threat to the swine industry. In this retrospective study, a phylogenetic analysis was conducted to address the epidemiology and evolutionary dynamics of this novel circovirus. The total positive sample rate of PCV3 was 26.7% (76/285) and has increased gradually over the past 3 years. Of these PCV3-positive samples, 22.3% (17/76) were coinfected with PCV2. PCV3 can be detected in multiple sample types with different positive rates, and the positive rate is highest among stillborn. We also divide PCV3 into three clades (PCV3a, PCV3b and PCV3c) based on two amino acid mutations (A24V and R27K) on the cap protein in this study. In addition, the origin of PCV3 was approximately 1966 and may have originated from a bat-associated circovirus. Our results suggested that PCV3 is widely distributed in southern China and has been circulating in swine herds for nearly half a century. PCV3 has evolved into different clades caused by mutations in cap proteins; thus, further research on PCV3 epidemiology should be conducted.

A community effort to protect genomic data sharing, collaboration and outsourcing

The human genome can reveal sensitive information and is potentially re-identifiable, which raises privacy and security concerns about sharing such data on wide scales. In 2016, we organized the third Critical Assessment of Data Privacy and Protection competition as a community effort to bring together biomedical informaticists, computer privacy and security researchers, and scholars in ethical, legal, and social implications (ELSI) to assess the latest advances on privacy-preserving techniques for protecting human genomic data. Teams were asked to develop novel protection methods for emerging genome privacy challenges in three scenarios: Track (1) data sharing through the Beacon service of the Global Alliance for Genomics and Health. Track (2) collaborative discovery of similar genomes between two institutions; and Track (3) data outsourcing to public cloud services. The latter two tracks represent continuing themes from our 2015 competition, while the former was new and a response to a recently established vulnerability. The winning strategy for Track 1 mitigated the privacy risk by hiding approximately 11% of the variation in the database while permitting around 160,000 queries, a significant improvement over the baseline. The winning strategies in Tracks 2 and 3 showed significant progress over the previous competition by achieving multiple orders of magnitude performance improvement in terms of computational runtime and memory requirements. The outcomes suggest that applying highly optimized privacy-preserving and secure computation techniques to safeguard genomic data sharing and analysis is useful. However, the results also indicate that further efforts are needed to refine these techniques into practical solutions.

Addressing Beacon re-identification attacks: quantification and mitigation of privacy risks

The Global Alliance for Genomics and Health (GA4GH) created the Beacon Project as a means of testing the willingness of data holders to share genetic data in the simplest technical context-a query for the presence of a specified nucleotide at a given position within a chromosome. Each participating site (or "beacon") is responsible for assuring that genomic data are exposed through the Beacon service only with the permission of the individual to whom the data pertains and in accordance with the GA4GH policy and standards.While recognizing the inference risks associated with large-scale data aggregation, and the fact that some beacons contain sensitive phenotypic associations that increase privacy risk, the GA4GH adjudged the risk of re-identification based on the binary yes/no allele-presence query responses as acceptable. However, recent work demonstrated that, given a beacon with specific characteristics (including relatively small sample size and an adversary who possesses an individual's whole genome sequence), the individual's membership in a beacon can be inferred through repeated queries for variants present in the individual's genome.In this paper, we propose three practical strategies for reducing re-identification risks in beacons. The first two strategies manipulate the beacon such that the presence of rare alleles is obscured; the third strategy budgets the number of accesses per user for each individual genome. Using a beacon containing data from the 1000 Genomes Project, we demonstrate that the proposed strategies can effectively reduce re-identification risk in beacon-like datasets.

Low copy numbers of FCGR3A and FCGR3B associated with Chinese patients with SLE and AASV

Low-affinity Fcγ receptors (FcγR) act as key mediators of the pathogenic effects of autoantibodies. In this study, we aimed to determine whether copy number variations (CNVs) in FCGR3A and FCGR3B were associated with systemic lupus nephritis (SLE) and ANCA-associated systemic vasculitis (AASV) in Chinese individuals. A total of 1118 individuals were enrolled, including 415 SLE patients, 139 AASV patients, and 564 healthy controls. FCGR3A and FCGR3B copy numbers (CNs) were determined by both a paralogue ratio test and TaqMan quantitative PCR assay. In the susceptibility associations, a low FCGR3B CN was significantly associated with SLE ( p = 5.01 × 10^-3; odds ratio (OR) 1.71; 95% confidence interval (CI) 1.17-2.48) and AASV ( p = 0.04; OR = 1.72; 95% CI 1.02-2.88). A low FCGR3A CN was also significantly associated with SLE ( p = 6.02 × 10^-3; OR 2.72; 95% CI 1.30-5.71) and AASV ( p = 0.042; OR 2.64; 95% CI 1.00-6.93). Further subphenotype analysis revealed that low CNs of FCGR3A and FCGR3B were significantly associated with clinical manifestations in SLE and AASV patients. Therefore, in this case-control study, we identified low CNs of FCGR2A and FCGR3B to be common risk factors for SLE and AASV.

Mosaicism in von Hippel-Lindau disease with severe renal manifestations

von Hippel-Lindau (VHL) disease is an inheritable multisystem tumor syndrome characterized by multiple benign and malignant tumors affecting multiple organs. VHL is the result of a germline mutation in the VHL tumor suppressor gene. Molecular genomic analysis routinely confirms the clinical diagnosis. However, the use of molecular diagnostic methods can often be insufficient for the detection of mosaic germline VHL mutations, making the diagnosis of some cases of VHL difficult. Here, we report the case of a VHL mosaic patient with bilateral renal lesions in the absence of other VHL-associated lesions. A VHL mutation was not originally detected by routine molecular testing. Nonetheless, the detection of a heterozygous c.194C>G (p.Ser65Trp) VHL mutation in the patient's daughter prompted further genetic assessment and eventually resulted in the finding of a mosaic c.194C>G (p.Ser65Trp) VHL mutation in the patient. The mutation rate was 18.8 ± 3.84% in peripheral leukocytes. As the frequency of VHL mosaicism remains underdetermined, the possibility of a diagnosis of mosaic VHL should be considered in patients with both typical and atypical VHL-associated manifestations.

MicroRNA expression patterns in the bovine mammary gland are affected by stage of lactation

The objective of this work was to determine the expression pattern of microRNA (miR) associated with cellular proliferation, lipid metabolism, and innate immunity in dairy cow mammary gland tissue at different stages of lactation. The expression of miR-10a, miR-15b, miR-16, miR-21, miR-31, miR-33b, miR-145, miR-146b, miR-155, miR-181a, miR-205, miR-221, and miR-223 was studied by real-time reverse-transcription PCR in tissue (n=7/stage) harvested via repeated biopsies during the dry period (-30 d prepartum), the fresh period (7 d postpartum), and early lactation (30 d postpartum). Except for miR-31, all miR studied increased in expression between the dry and fresh periods. Among those upregulated, the expression of miR-221 increased further at early lactation, suggesting a role in the control of endothelial cell proliferation or angiogenesis, whereas the expression of miR-223 decreased at early lactation but to a level that was greater than in the dry period, suggesting it could play a role in the mammary response to pathogens soon after parturition. The expression of miR-31, a hormonally regulated miR that inhibits cyclin gene expression, was greater at early lactation compared with the dry period. From a metabolic standpoint, the consistent upregulation of miR-33b during early lactation compared with the dry period suggests that this miR may exert some control over lipogenesis in mammary tissue. Overall, results indicate that expression of miR associated with transcriptional regulation of genes across diverse biological functions is altered by stage of lactation. The specific roles of these miR during lactation will require further research.

New mutations in the transglutaminase 1 gene in three families with lamellar ichthyosis

BACKGROUND

Autosomal recessive lamellar ichthyosis (LI) is a severe skin disorder characterized by generalized hyperkeratosis. Gene mutation in transglutaminase 1 (TGM1), which mediates cross-links in the formation of the cell envelope during terminal differentiation of epidermis, has been identified as a cause of LI.

OBJECTIVES

To determine mutations of TGM1 gene in three Chinese families with LI.

METHODS

The TGM1 gene was sequenced to identify disease-causing mutations in the three families with LI. One of the results was confirmed by using reverse transcriptase PCR and in situ hybridization. An in situ transglutaminase (TGase) 1 assay was performed to estimate TGase 1 activity in the patients' skin.

RESULTS

Four novel mutations of keratinocyte TGase1 (Q203X, D254N, R687H and IVS4 + 1G-->T) were found in the three families. No TGase 1 mRNA was detected in patient skin using RT-PCR and in situ hybridization, and the in situ TGase assay showed that there was no or decreased TGase 1 activity in patient skin.

CONCLUSIONS

Our findings suggest that four novel mutations in TGM1 gene result in decrease or absence of TGase activity in the skin and, as a consequence, cause the phenotype of LI.

Obesity, insulin resistance and oxidative stress: implication for breast carcinogenesis

Abstract #6025

Background: Obesity is associated with a modest increase in the risk of postmenopausal breast cancer (RR = 1.5 – 2.0); but, because nearly a third of the U.S. population is obese, the population attributable risk is estimated at 15%. Obesity can cause insulin resistance culminating in Type II diabetes. Notably, breast cancer incidence is significantly increased in the years preceding a diagnosis of type II diabetes. Because IGFBP-1 expression is tightly regulated by insulin, it is an excellent marker of insulin resistance in healthy individuals.
 Methods: These data are based on well-annotated prospectively acquired baseline blood and breast tissue samples from 72 high risk women between the ages of 37 and 86 years who participated in a chemoprevention trial. None of the women had been diagnosed with Type II diabetes. Women with plasma IGFBP1 levels in the lowest tertile (mean 2.1 ng/ml) were classified as insulin-resistant.
 Results: Plasma IGFBP1 was strongly inversely correlated with BMI (R2 = 0.247, P < 0.0001). Insulin-resistant women had marginally higher mean plasma free estradiol levels than women not classified as insulin-resistant (2.13 x 10-12M versus 1.53 x 10-12M, P = 0.072). There was no difference in plasma IGF1, IGF2, or IGFBP3 levels. Illumina whole genome expression microarray data was available for breast tissue from 55 women. Women classified as insulin-resistant showed evidence of an adaptive response to oxidative stress based on significant upregulation of NQO1, GSTK1, CYP4ZP2, and SRXN1 (P < 0.001).
 Conclusions: Marginally increased circulating estradiol may contribute to the increased breast cancer risk observed in insulin-resistant women. However, insulin resistance increases oxidative stress in breast tissue and may promote carcinogenesis through induction of oxidative DNA damage.

Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 6025.

Cell-based breast cancer risk stratification based on DNA methylation in fine needle aspiration samples

1508

Background: Tumor suppressor gene (TSG) methylation is identified in nearly all breast cancers, but rarely in histologically normal breast tissue from wonen unaffected with breast cancer. Its occurrence in high risk preneoplasia and in benign breast tissue adjacent to breast cancer suggests that it may represent a high risk field change that could be exploited for cell-based breast cancer risk stratification. Methods: TSG methylation was measured by quantitative methylation-specific real time PCR in 53 breast tumor fine needle aspiration (FNA) biopsies, 84 cellular random periareolar FNAs (RP-FNA) ipsilateral or contralateral to these cancers, 36 cellular RP- FNAs from unaffected women at high risk for breast cancer by the Gail model, and 95 cellular RP-FNAs from unaffected women at lower risk by the Gail model. Results: The breast tumors showed a high frequency of TSG methylation: RASSF1A 80%, HIN-1 65%, Cyclin D2 60%, RAR-β2 53%, and APC 47%. In general, RP-FNA samples from cancer patients and Gail high risk patients showed a greater frequency of methylation than samples from Gail lower risk patients: RASSF1A 43% vs. 21%, P = 0.001, HIN-1 32% vs. 20%, P = 0.05; Cyclin D2 18% vs. 9%, P = 0.10; RAR-β2 21% vs. 18%, P = 0.68; and APC 25% vs. 16%, P = 0.17. Twelve of 215 RP-FNA samples (5%) showed very high levels of methylation (>10% methylation for two or more genes). Only two of these samples were from women classified as lower risk by the Gail model. Methylation frequencies were entirely independent of cell yields but the frequency of RASSF1A methylation increased with increasing Masood scores (P = 0.05). Methylation of RASSF1A in one breast was highly predictive of RASSF1A methylation in the opposite breast (P < 0.0001). Conclusions: TSG methylation appears to be a breast cancer risk-associated field change that can be quantified in RP-FNA samples. RASSF1A methylation occurs frequently in benign breast epithelium, provides reasonable discrimination between high and lower risk breasts (O.R. = 2.0), is related to cytological atypia, and may be an early marker of a methylator phenotype. Quantification of TSG methylation in RP-FNA samples may provide a valuable surrogate endpoint biomarker for Phase II prevention trials.

No significant financial relationships to disclose.

Assessment of tumor suppressor gene methylation for breast cancer risk screening

1004

Background: Tumor suppressor gene (TSG) methylation is frequently detected in benign proliferative breast tissue suggesting that it occurs early in breast carcinogenesis. If it can be screen-detected and is associated with breast cancer risk it could be exploited for breast cancer prevention. Methods: Nipple duct lavage (NDL) samples, obtained from 150 women selected to represent a wide range of breast cancer risk, were evaluated by quantitative methylation-specific real time PCR. High risk breasts were defined as those contralateral to a breast cancer (N = 63) and those of women with a 5-year Gail risk ≥ twice the age- and race-matched general population risk (N = 64). The prevelence of TSG methylation and marked atypia was compared for high risk and lower risk breasts using Chi-square. Data for breasts ipsilateral to a breast cancer are shown for comparison, but not included in the calculations for the high risk category. Results: Samples with adequate cellularity were obtained for 219 breasts (76%). The proportion of healthy breasts with ≥ 1% of the gene copies methylated was 13% for Cyclin D2, 19% for APC, 19% for HIN-1, 16% for RASSF1A, and 9% for RAR-beta. RAR-beta provided the best risk discrimination as 15% of high risk breasts were methylated at a level that exceeded the 95th percentile of the lower risk breasts (0.9% of gene copies methylated, P = 0.05). For the table , methylation fractions for all five genes were summed and the threshold for classifying a breast as positive was set to the 95th percentile of the lower risk breasts (methylation sum = 25.0%). Both methylation and marked atypia provide some discrimination between high and lower risk breasts; the combination, however, provides the best discrimination (24% marker positive for high risk versus 9% for lower risk, P = 0.02). Conclusions: TSG methylation in NDL samples is a marker of breast cancer risk that is complementary to cytology.

[Table: see text]

[Table: see text]

Mutations in SCN9A, encoding a sodium channel alpha subunit, in patients with primary erythermalgia

Primary erythermalgia is a rare autosomal dominant disease characterised by intermittent burning pain with redness and heat in the extremities. A previous study established the linkage of primary erythermalgia to a 7.94 cM interval on chromosome 2q, but the causative gene was not identified. We performed linkage analysis in a Chinese family with primary erythermalgia, and screened the mutations in the two candidate genes, SCN9A and GCA, in the family and a sporadic patient. Linkage analysis yielded a maximum lod score of 2.11 for both markers D2S2370 and D2S2330. Based on critical recombination events in two patients in the family, we further limited the genetic region to 5.98 cM between D2S2370 and D2S2345. We then identified two missense mutations in SCN9A in the family (T2573A) and the sporadic patient (T2543C). Our data suggest that mutations in SCN9A cause primary erythermalgia. SCN9A, encoding a voltage-gated sodium channel alpha subunit predominantly expressed in sensory and sympathetic neurones, may play an important role in nociception and vasomotor regulation.

DNA-based prenatal diagnosis in a Chinese family with xeroderma pigmentosum group A

BACKGROUND

Xeroderma pigmentosum (XP) is a group of autosomal recessive diseases characterized by hypersensitivity to ultraviolet rays. Among its eight complementation groups, XP group A (XPA) is the most severe type. The XPAC gene has been identified as the defective gene in XPA patients.

OBJECTIVES

To examine genomic DNA from a Chinese family with XPA, to determine the XPAC mutation and, after genetic counselling, to undertake DNA-based prenatal diagnosis in a subsequent pregnancy.

METHODS

Fetal DNA was extracted from amniotic fluid and used to amplify exon 5 of XPAC containing the potential mutation. Direct sequencing and restriction endonuclease digestion were used for prenatal diagnosis.

RESULTS

We identified a homozygous nonsense XPAC mutation of 631C-->T, which results in an R211X mutation in XPA protein, in the proband. Both her parents are heterozygous. Prenatal diagnosis demonstrated a heterozygous sequence predicting an unaffected child, and a healthy girl was born.

CONCLUSIONS

These data provide the first example of a DNA-based prenatal test for genodermatosis in China.

Immunobiology and long-term graft function in a transplant heterotopic renal rat model

BACKGROUND

The Th1-Th2 paradigm proposes clonal expansion of Th2 lymphocytes as the basis of allograft tolerance. The Th2 cells have been found to be present in recipients with long-term allograft survival. However, the presence of Th2 cells and tolerance is not a uniform finding. Previously we have shown that pre-engraftment single dose rapamycin and a 7-d course of cyclosporin induce transplantation tolerance to 120 d. In the present study, we investigated the immunobiology of grafts in a long-term follow-up (>350 d).

METHODS

Kidney allografts (n = 7), isografts (n = 5) and single nephrectomy (n = 3) groups were followed for 350 +/- 87 d. Heterotopic kidney transplant was performed by the same surgeon in the allograft group (ACI-Lewis) and the isograft group (Lewis-Lewis). The left kidney was removed in the single nephrectomy group. The allograft group was treated with pre-engraftment single dose rapamycin and a 7-d course of cyclosporin. A kidney biopsy was performed at midpoint time for histological study and tissue was frozen for measuring intragraft cytokine expression (IL-4, IL-10) in all animals. Prior to biopsy, serum blood urea nitrogen (BUN) and creatinine (Cr) levels were studied. Serum BUN, Cr levels, plus 24-h urinary protein (PRO) were measured prior to sacrifice. Randomly, four allograft rats received skin grafts (ACI, Lewis and Buffalo skin donors) after kidney biopsy. Skin grafts were studied for a mean of 6 weeks for signs of acceptance or rejection. Analysis of variance (ANOVA) with Tukey's test was used; p < 0.05 was considered statistically significant.

RESULTS

The mean follow up was 352 +/- 87 d. BUN and Cr levels at biopsy time (mean 214 d) were not statistically different between the three groups (p = 0.19 and p = 0.66). At sacrifice (mean 352 d), BUN, Cr and PRO were statistically different between allograft and isograft groups (p = 0.013), and between allograft and single nephrectomy groups (p = 0.027). Functional and histological signs of graft loss occurred in three of seven (42.8%) of the allografts at 352 d. Using BANFF criteria, three allografts at biopsy time and seven allografts (100%) and four isografts (80%) at sacrifice time developed chronic histologic changes. Intragraft overexpression of IL-4 and IL-10 was seen at biopsy and sacrifice time in six of seven allografts and one of five isografts. All donor specific skin grafts (ACI-Lewis) on allografts were accepted and third party (Buffalo) donor skin grafts were rejected in all animals (>95% skin necrosis).

CONCLUSIONS

This highly stringent, functional, renal transplant model yields 100% normal renal function as compared with isografts at 120 d follow-up. With the follow-up extended to 350 d, 43% of the allografts loose function and develop a chronic allograft histology despite a demonstrated intragraft Th2 cytokine dominance and donor specific skin graft acceptance.

[A novel mutation of the ALAS2 gene in a family with X-linked sideroblastic anemia]

OBJECTIVE

To confirm the mutation of ALAS2 gene is the cause of sideroblastic anemia in a family.

METHODS

Polymerase chain reaction (PCR) was used to amplify the microsatellite DXS 991, DXS 1199 in the chromosome Xp11.22 linked gene ALAS2 and haplotype analysis was performed in a kindred with 2 patients and 7 normal members. All cDNA encoded regions in the ALAS2 gene of the patients and their normal siblings were cloned, sequenced and compared.

RESULT

Both brother patients had the same allele of ALAS2 and their normal siblings did not. The mutation in the patients' ALAS2 gene was exon 5 A523G, causing threonine to alanine; and exon 3 T372C, leucine to proline. The latter located in the splicing region, its significance is not clear.

CONCLUSION

The pathogenesis of this kindred of X-linked sideroblastic anemia (XLSA) involved a novel mutation in ALAS2 exon 5.

Defining hepatocellular chimerism in a liver failure patient bridged with hepatocyte infusion

BACKGROUND

A practical method of monitoring engraftment by transplanted hepatocytes for the purpose of bridging human liver failure to native regeneration is described.

METHODS

A previously healthy 37-year-old female with a 2-week history of a febrile illness presented with fulminant liver failure. Findings on admission included the following: illicit drug use, serum hepatitis B surface antigen positive, grade 1 encephalopathy, prothrombin time (pt) >100 sec, F7<1%, NH3 150 micromol/L, alanine aminotransferase 4079 U/L, total bilirubin level 11.4 mg/dl, and glucose 70 mg/dl (on IV D10). With immunosuppression, 8.8x10(8), 96% viable human hepatocytes were intraportally infused. Clinical chemistries, total sHLA class I, and ELISA to measure donor-specific sHLA-A1 and -B8 were recorded. Serial transjugular liver biopsies were performed and pooled for histological examination, DNA extraction, and HLA DNA typing.

RESULTS

The patient fully recovered. At months 3 and 4 with donor biopsy specimen class I HLA DNA no longer detectable, immunosuppression was tapered off. The patient is clinically normal, serum hepatitis B surface antigen negative at 10 months of follow-up.

CONCLUSIONS

Bridging liver failure with donor hepatocytes with HLA class I antigen disparate from recipients is clinically feasible, and allows for a marker, combined with serial graft histology, to safely wean immunosuppression when native liver regeneration succeeds.

Spatial-specific TGF-beta1 adenoviral expression determines morphogenetic phenotypes in embryonic mouse lung

The precise spatial-temporal role that expression and activation of transforming growth factor (TGF)-beta plays in mammalian organ morphogenesis remains incompletely understood. Using replication deficient adenoviral vectors containing engineered TGF-beta1 cDNAs, we studied the spatial effects of locally over-expressing either latent or mutated, constitutively active TGF-beta1 protein during embryonic mouse lung branching morphogenesis in culture. Transfer of exogenous genes into lung epithelium was achieved by intra-tracheal micro-injection of recombinant adenovirus, while submerging lungs in virus resulted in gene transfer into the pleura and subjacent mesenchymal cells, as revealed by cytochemical staining for beta-galactosidase. Only lungs transfected with active, but not latent TGF-beta1 gene, showed elevated levels of active TGF-beta. Epithelial over-expression of active, but not latent TGF-beta1, via intra-tracheal micro-injection inhibited lung branching morphogenesis by 36 %. In contrast, lungs submerged with either active or latent TGF-beta1 recombinant virus did not demonstrate an inhibitory effect upon branching. Pulmonary gene regulation was assayed by competitive polymerase chain reaction coupled with reverse transcription. Direct respiratory tract micro-injection of adenovirus over-expressing active TGF-beta1 resulted in a dose-dependent inhibition of epithelial surfactant protein (SP)-C and SP-B mRNA levels by up to 76 % and 70 %, respectively, while in contrast, fibronectin and matrix Gla protein (MGP) mRNA levels remained stable. However, lungs that had been submerged in adenovirus expressing active TGF-beta1 demonstrated a concentration-dependent induction of both fibronectin and MGP mRNA levels up to 4.3- and 4.7-fold respectively in the presence of 1 x 10(11) pfu/ml active TGF-beta1 virus. On the other hand, lungs treated with adenovirus expressing latent TGF-beta1 either by micro-injection or submerging failed to demonstrate any regulatory effect either upon epithelial or mesenchymal gene expression. We conclude that adenovector-mediated over-expression of activated TGF-beta1 in specific spatial compartments results respectively in either inhibition of branching morphogenesis and epithelium-specific gene expression, or in induction of matrix gene expression without affecting morphogenesis or epithelium-specific gene expression, depending on the route of administration. Also, the lack of effect of latent TGF-beta1 over-expression strongly suggests that TGF-beta activation per se provides an important locus of fine regulation of the spatial effects of TGF-beta signaling during embryonic lung branching morphogenesis.

[Pathogenic gene linkage analysis and hemopoietic characteristics in a kindred with sideroblastic anemia]

OBJECTIVE

Analysis of pathogenic gene linkage and hemopoietic characteristics in a kindred with sideroblastic anemia.

METHODS

PCR amplification of the microsatellites DXS991,DXS1199 in chromosome Xp11.22 linked gene ALAS2 and of the microsatellite DXS1226 in Xp22. 13 linked another irrelevant gene and analysis of gene linkage in a kindred with 2 patients and 7 normal persons. The bone marrow hemopoietic cells from 2 patients were cultured in condition culture matrix with various cytokines added in and the CFU-E, CFU-GM and CFU-Meg formations were observed at different times.

RESULTS

The kindred study revealed that pathogenic gene linked with DXS991 and DXS1199 but did not link with DXS1226.Hemopoietic cell culture showed that erythroid colonies of the two patients grew more vigorously than controls and they could grew in the absence of Epo except in common condition matrix. The erythroid colonies withered after a week and were smaller than the controls after 13 days.

CONCLUSION

The kindred is subject to an X-linked sideroblastic anemia(XLSA) with the pathogenic gene ALAS2 involved. In XLSA,the function of stem cells is primarily normal before erythropoiesis, then the erythroid progenitors become dysplasia.

Intragraft cytokine expression in tolerant rat renal allografts with rapamycin and cyclosporin immunosuppression

The Th-1/Th-2 paradigm proposes clonal expansion of Th-2 lymphocytes as the basis of tolerance towards allografts. Intragraft cytokine expression was evaluated in a highly stringent model of renal transplantation. ACI and Lewis rats were used as donors and recipients, respectively, for heterotopic renal transplantation. Group A (n = 8) received a single dose of rapamycin and cyclosporin 12 h prior to engraftment, followed by 7 d of cyclosporin post-operatively. Isografts (Group B, n = 5) and control allografts (Group C, n = 4) received no immunosuppression. Sacrifice was performed after 120 d. Intragraft expression of IL-10, IL-4, and IFN-gamma was determined using qualitative reverse transcriptase-polymerase chain reaction (RT-PCR). All groups had functionally normal grafts at sacrifice, with 50% histological tolerance among Group A animals. No isografts showed evidence of cellular infiltrate, and all control allografts showed severe rejection. IL-10 was only detected in the tolerant animals (p < 0.001). Similarly, IL-4 was detected predominantly in the tolerant allografts (p < 0.05). IFN-gamma was only isolated in rejected allografts, whether treated or untreated (p < 0.001). We conclude that the expansion of Th-2 cells is associated with tolerance, while the expansion of Th-1 cell is associated with acute cellular rejection.

Intragraft cytokine expression and tolerance induction in rat renal allografts

BACKGROUND

Intragraft cytokine expression was evaluated in a model of renal transplantation. ACI and Lewis rats were used as donors and recipients, respectively, for heterotopic renal transplantation.

METHODS

Treated allograft rats (n=10) received a preoperative dose of rapamycin and cyclosporine, followed by 7 days of cyclosporine postoperatively. Isograft rats (n=5) and control allograft rats (n=4) received no immunosuppression. Sacrifice was performed after 120 days. Expression of interleukin (IL)-4, IL-10, and interferon-gamma (IFN-gamma) transcripts was determined with semiquantitative reverse transcriptase-polymerase chain reaction.

RESULTS

All treated allograft rats had normal function with 50% histologic rejection. All isografts had normal function. IL-4 and IL-10 were in greater density in allografts with normal histology, whereas IFN-gamma was only seen in allografts with cellular rejection. No IL-10 was seen in isografts, but IL-4 was detected in 3/5 isografts.

CONCLUSIONS

We conclude that the lymphocyte population's elaboration of IL-4 and IL-10 is associated with tolerance, whereas the production of IFN-gamma and absence of IL-4 is associated with histology suggestive of acute cellular rejection.

Impaired lung branching morphogenesis in the absence of functional EGF receptor

The mammalian lung develops through branching morphogenesis which is controlled by growth factors, hormones, and extracellular matrix proteins. We have evaluated the role of EGF-receptor signaling in lung morphogenesis by analyzing the developmental phenotype of lungs in mice with an inactivated the EGF-receptor gene both in vivo and in organ culture. Neonatal EGF-receptor-deficient mice often show evidence of lung immaturity which can result in visible respiratory distress. The lungs of these mutant mice had impaired branching and deficient alveolization and septation, resulting in a 50% reduction in alveolar volume and, thus, a markedly reduced surface for gas exchange. The EGF-receptor inactivation also resulted in type II pneumocyte immaturity, which was apparent from their increased glycogen content and a reduced number of lamellar bodies. The defective branching was already evident at Day 12 of embryonic development. When explants of embryonic lungs from Day 12 embryos were cultured under defined conditions, the branching defect in EGF-receptor-deficient lungs was even more pronounced, with only half as many terminal buds as normal lungs. EGF treatment stimulated the expression of surfactant protein C and thyroid transcription factor-1 in cultured normal lungs, but not in EGF-receptor-deficient lungs, suggesting that EGF-receptor signaling regulates the expression of these marker genes during type II pneumocyte maturation. Taken together, our data indicate that signal transduction through the EGF receptor plays a major role in lung development and that its inactivation leads to a respiratory distress-like syndrome.

Abrogation of transforming growth factor-beta type II receptor stimulates embryonic mouse lung branching morphogenesis in culture

TGF-beta1 is a known inhibitor of branching morphogenesis when added exogenously to mouse embryonic lungs in culture. However, the issue of whether endogenous TGF-beta signaling has a function in the process of lung organogenesis is not completely resolved. We utilized immunoperturbation and antisense oligodeoxynucleotide inhibitory strategies to abrogate TGF-beta type II receptor function in embryonic mouse lungs undergoing branching morphogenesis in serumless explant culture. Antisera directed against a TGF-beta type II receptor N-terminal peptide that perturbs TGF-beta ligand-receptor binding increased branching by 70%. Similarly, antisense TGF-beta type II receptor oligodeoxynucleotides (40 microM) resulted in a 58% increase in branching, compared to scrambled and mismatched sequence controls, while TGF-beta, type II receptor mRNA and its protein expression levels were suppressed by 95 and 84%, respectively. Addition of exogenous TGF-beta1 did not overcome the stimulatory effects either of TGF-beta type II receptor immunoperturbation or of antisense oligodeoxynucleotide treatment on lung branching morphogenesis. Using in situ hybridization and immunohistochemistry, both TGF-beta type II receptor mRNA and protein were localized to the epithelium lining the developing airways, and to the surrounding mesenchyme, indicating that TGF-beta type II receptor is an important regulator of epithelial-mesenchymal interaction. Exogenous TGF-beta1 decreased cyclin A mRNA levels in control embryonic lung explants, while TGF-beta type II receptor antisense oligodeoxynucleotides prevented the downregulation of cyclin A mRNA expression by exogenous TGF-beta1. In addition, PCNA immunostaining of the primitive bronchial epithelium was increased in the presence of TGF-beta type II receptor antisense oligodeoxynucleotides either alone or together with exogenous TGF-beta1, whereas TGF-beta1 alone decreased PCNA staining. Thus, abrogation of TGF-beta type II receptor expression prevented TGF-beta1-induced epithelial cell G1 arrest. These results demonstrate, for the first time, that abrogation of the TGF-beta type II receptor stimulates embryonic lung organogenesis in culture and reverses the negative influence of endogenous TGF-beta signaling upon epithelial cell cycle progression.

Cytokine-stimulated astrocytes damage human neurons via a nitric oxide mechanism

Astrocytes have been reported to play a neuropathogenic role within the brain, although little is known about the mechanism underlying astrocyte-mediated neuronal injury. We investigated the hypothesis that cytokine-stimulated astrocytes adversely affect neuronal cell survival via generation of the free radical nitric oxide (NO). Primary human astrocytes produced substantial amounts of NO in response to interleukin (IL)-1 alpha or IL-1 beta, which was blocked by the NO synthase inhibitor NG-mono-methyl-L-arginine (NMMA). IL-1 beta-induced NO production was markedly potentiated by interferon (IFN)-gamma. IL-1 receptor agonist protein (IRAP) totally blocked NO generation by cytokine-stimulated astrocytes. Using reverse transcription-polymerase chain reaction and sequencing analyses of the astrocyte NO synthase gene, we found a single band encoding for a 615 bp product that was identical to the corresponding sequence reported for human hepatocytes. Treatment of human fetal brain cell cultures with IL-1 beta plus IFN-gamma resulted in marked neuronal loss, as assessed by microscopic analysis and measurement of lactate dehydrogenase release. This cytokine-induced neuronal damage was blocked by simultaneous treatment of the brain cell cultures with NMMA or IRAP, suggesting a critical role of IL-1. These findings indicate that cytokine-stimulated astrocytes are neurotoxic via a NO-mediated mechanism and point to potential new therapies for neurodegenerative disorders that involve cytokines and reactive astrocytes.

TTF-1 regulates lung epithelial morphogenesis

TTF-1 is a homeodomain transcriptional factor expressed in thyroid, lung, and parts of the brain. In vitro, TTF-1 can activate the promoter of thyroid- and pulmonary-specific genes. We postulated that TTF-1 not only is essential for the activation of tissue-specific genes, but also may directly participate in epithelial cell morphogenesis. To test this postulate, we used an antisense oligonucleotide inhibitory strategy in an in vitro model of embryonic mouse lung branching morphogenesis. This strategy suppressed TTF-1 translation and inhibited lung branching morphogenesis. The resulting abnormal phenotype was characterized by hyperplastic and unorganized proliferation of epithelial cells in the airways. The mesenchymal compartment of the lung appeared to be unaffected. These results demonstrate, for the first time, that the expression of a homeoprotein transcriptional regulator is necessary for lung epithelial morphogenesis.

Abnormal lung development and cleft palate in mice lacking TGF-beta 3 indicates defects of epithelial-mesenchymal interaction

A broad spectrum of biological activities has been proposed for transforming growth factor-beta 3 (TGF-beta 3). To study TGF-beta 3 function in development, TGF-beta 3 null mutant mice were generated by gene-targeting. Within 20 hours of birth, homozygous TGF-beta 3-/- mice die with unique and consistent phenotypic features including delayed pulmonary development and defective palatogenesis. Unlike other null mutants with cleft palate, TGF-beta 3-/- mice lack other concomitant craniofacial abnormalities. This study demonstrates an essential function for TGF-beta 3 in the normal morphogenesis of palate and lung, and directly implicates this cytokine in mechanisms of epithelial-mesenchymal interaction.

Characterization and response to interleukin 1 and tumor necrosis factor of immortalized murine biliary epithelial cells

BACKGROUND & AIMS

Biliary epithelial cells are the target of numerous immune-mediated liver diseases, yet their role in pathogenesis remains unclear because of difficulties in obtaining pure preparations. The aim of this study was to establish pure clones of immortalized murine intrahepatic biliary epithelial cells.

METHODS

The transgenic mouse harboring the SV40 thermosensitive immortalizing mutant gene TsA58 under the control of the major histocompatibility complex class I promoter was used to establish conditionally immortalized intrahepatic bile duct cells by countercurrent centrifugal elutriation and clonal dilution.

RESULTS

Immortalized clones of cells expressing cytokeratin 19, which organized themselves into ductlike structures, were obtained. On electron-microscopic sections, cells were well differentiated and polarized. Cells proliferate in response to epidermal growth factor, interleukin 1 alpha, and tumor necrosis factor alpha. Using the reverse-transcriptase polymerase chain reaction technique, these cells were found to contain messenger RNA, which encodes for the interleukin 1 and tumor necrosis factor receptors.

CONCLUSIONS

The availability of unlimited numbers of pure bile duct cells that behave in an identical fashion to biliary epithelial cells from "normal" mice will allow for more rigorous studies of the behavior and function of this epithelium.

Biophysical characterization of one-, two-, and three-tandem repeats of human mucin (muc-1) protein core

Until recently mucin tandem repeat protein cores were believed to exist in random-coil conformations and to attain structure solely by the addition of carbohydrates to serine and threonine residues. Matsushima et al. (Proteins Struct. Funct. Genet., 7: 125-155, 1990) recently proposed a model of the secondary structure of proline rich tandem repeat proteins that has challenged this idea, especially for the case of the human polymorphic epithelial mucin encoded by the muc-1 gene. We report here results of structural analyses of the muc-1 protein core by using synthetic peptide analogues. Synthetic peptides were prepared to correspond to one-, two-, and three-tandem repeats of muc-1. Results of one- and two-dimensional 1H NMR correlation spectroscopy on these peptides confirm that the muc-1 protein core is not a random-coil secondary structure. Long-lived amide protons are protected in D2O, and increasing spectral complexity in the region of the beta-protons of Asp2 and His 15 reveals that structural changes are occurring as the number of repeats increases. The greatest changes occur when the number of repeats increases from one to two. These results are supported by the reactivity of a panel of monoclonal antibodies raised against tumor associated muc-1 with these synthetic peptides in enzyme-linked immunosorbent assay. The primary immunodominant mucin epitope, PDTRP, does not appear to attain a native conformation in the single repeat peptide (20 amino acids, starting with P), but is expressed on peptides with multiple repeats. Intrinsic viscosity measurements of the peptide containing three repeats indicate that an ordered structure present in solution is rod shaped. The circular dichroism spectrum of the same peptide is dominated by proline in the trans conformation. These results are all consistent with the prediction that the muc-1 tandem repeat polypeptide core forms a polyproline beta-turn helix.

Recombinant vaccinia mucin vector: in vitro analysis of expression of tumor-associated epitopes for antibody and human cytotoxic T-cell recognition

We have constructed a recombinant vaccinia virus vector that contains human mucin MUC-1 cDNA. Analysis of the recombinant virus isolates showed the tendency of the vaccinia to delete large portions of the mucin tandem repeat region. Epstein-Barr virus (EBV)-immortalized B cell lines from humans and chimpanzees were infected and analyzed for expression of the mucin on the cell surface and the presence of specific epitopes in the tandem repeat region previously shown to be preferentially expressed on tumor cells and recognized by tumor-specific mouse monoclonal antibodies and human cytotoxic T lymphocytes (CTL). We found that this recombinant vector encodes expression of mucin that contains all the epitopes recognized by the antibodies. The tumor-specific epitopes can be further exposed by inhibition of O-linked glycosylation in infected cells. Lack of multiple tandem repeats, however, prevents major histocompatibility complex (MHC)-unrestricted recognition by the CTL of the majority of infected cell lines. Still, we show two examples of an apparent MHC-restricted recognition of vaccinia-encoded mucin that may depend on one or very few rare human lymphocyte antigen (HLA) types capable of presenting the mucin peptides.

Expression of tumor-associated epitopes on Epstein-Barr virus-immortalized B-cells and Burkitt's lymphomas transfected with epithelial mucin complementary DNA

Mucins are among the best described human tumor-associated antigens. At least 73 tumor-reactive anti-mucin antibodies have been described; in addition, we have previously demonstrated the existence of tumor-specific cytotoxic T-lymphocyte epitopes on the mucin produced by breast and pancreatic tumors. To determine whether the appearance of tumor-associated mucin epitopes can be explained by altered post-translational modification of mucin in tumors, or whether the generation of these epitopes requires changes in the mucin gene itself, we studied four Burkitt's lymphomas and six Epstein-Barr virus-immortalized B-cell lines transfected with an expression construct containing the mucin complementary DNA. Transfected cell lines showed stable maintenance of the mucin gene, which comprises 20 or more tandem repeats of a 60-nucleotide sequence. Transfected cells expressed many tumor-associated mucin epitopes, suggesting that the changes in mucin synthesis seen in breast and pancreatic tumors are present in other malignant cell types as well. Furthermore, even though each cell line was transfected with the identical mucin construct, each expressed a different subset of tumor-associated mucin epitopes. This suggests that the specificity of these epitopes for tumors is not due to genetic alterations of the mucin gene in tumors. Incubating transfected cells with phenyl-N-acetyl-alpha-galactosaminide, an inhibitor of O-linked glycosylation, altered cell surface carbohydrate structures and resulted in increased expression of all tumor-associated epitopes, implicating incomplete glycosylation of mucin in the generation of these epitopes. These findings suggest that alterations in the posttranslational modification of normal gene products can result in the expression of novel epitopes. Furthermore, the ability to transfect cancer patients' Epstein-Barr virus-immortalized B-cell lines with mucin will provide an unlimited supply of autologous, mucin-bearing cells with which to study these patients' T-cell response to mucin.