Multimodal profiling reveals tissue-directed signatures of human immune cells altered with age

The immune system comprises multiple cell lineages and heterogeneous subsets found in blood and tissues throughout the body. While human immune responses differ between sites and over age, the underlying sources of variation remain unclear as most studies are limited to peripheral blood. Here, we took a systems approach to comprehensively profile RNA and surface protein expression of over 1.25 million immune cells isolated from blood, lymphoid organs, and mucosal tissues of 24 organ donors aged 20-75 years. We applied a multimodal classifier to annotate the major immune cell lineages (T cells, B cells, innate lymphoid cells, and myeloid cells) and their corresponding subsets across the body, leveraging probabilistic modeling to define bases for immune variations across donors, tissue, and age. We identified dominant tissue-specific effects on immune cell composition and function across lineages for lymphoid sites, intestines, and blood-rich tissues. Age-associated effects were intrinsic to both lineage and site as manifested by macrophages in mucosal sites, B cells in lymphoid organs, and T and NK cells in blood-rich sites. Our results reveal tissue-specific signatures of immune homeostasis throughout the body and across different ages. This information provides a basis for defining the transcriptional underpinnings of immune variation and potential associations with disease-associated immune pathologies across the human lifespan.

Cross-tissue immune cell analysis reveals tissue-specific features in humans

Despite their crucial role in health and disease, our knowledge of immune cells within human tissues remains limited. We surveyed the immune compartment of 16 tissues from 12 adult donors by single-cell RNA sequencing and VDJ sequencing generating a dataset of ~360,000 cells. To systematically resolve immune cell heterogeneity across tissues, we developed CellTypist, a machine learning tool for rapid and precise cell type annotation. Using this approach, combined with detailed curation, we determined the tissue distribution of finely phenotyped immune cell types, revealing hitherto unappreciated tissue-specific features and clonal architecture of T and B cells. Our multitissue approach lays the foundation for identifying highly resolved immune cell types by leveraging a common reference dataset, tissue-integrated expression analysis, and antigen receptor sequencing.

Therapeutically expanded human regulatory T-cells are super-suppressive due to HIF1A induced expression of CD73

The adoptive transfer of regulatory T-cells (Tregs) is a promising therapeutic approach in transplantation and autoimmunity. However, because large cell numbers are needed to achieve a therapeutic effect, in vitro expansion is required. By comparing their function, phenotype and transcriptomic profile against ex vivo Tregs, we demonstrate that expanded human Tregs switch their metabolism to aerobic glycolysis and show enhanced suppressive function through hypoxia-inducible factor 1-alpha (HIF1A) driven acquisition of CD73 expression. In conjunction with CD39, CD73 expression enables expanded Tregs to convert ATP to immunosuppressive adenosine. We conclude that for maximum therapeutic benefit, Treg expansion protocols should be optimised for CD39/CD73 co-expression.

Evidence that Cd101 is an autoimmune diabetes gene in nonobese diabetic mice

We have previously proposed that sequence variation of the CD101 gene between NOD and C57BL/6 mice accounts for the protection from type 1 diabetes (T1D) provided by the insulin-dependent diabetes susceptibility region 10 (Idd10), a <1 Mb region on mouse chromosome 3. In this study, we provide further support for the hypothesis that Cd101 is Idd10 using haplotype and expression analyses of novel Idd10 congenic strains coupled to the development of a CD101 knockout mouse. Susceptibility to T1D was correlated with genotype-dependent CD101 expression on multiple cell subsets, including Foxp3(+) regulatory CD4(+) T cells, CD11c(+) dendritic cells, and Gr1(+) myeloid cells. The correlation of CD101 expression on immune cells from four independent Idd10 haplotypes with the development of T1D supports the identity of Cd101 as Idd10. Because CD101 has been associated with regulatory T and Ag presentation cell functions, our results provide a further link between immune regulation and susceptibility to T1D.

NKG2D-RAE-1 receptor-ligand variation does not account for the NK cell defect in nonobese diabetic mice

NK cells from NOD mice induced with poly(I:C) in vivo exhibit low cytotoxicity against a range of target cells, but the genetic mechanisms controlling this defect are yet to be elucidated. Defects in the expression of NKG2D and its ligands, the RAE-1 molecules, have been hypothesized to contribute to the reduced NK function present in NOD mice. In this study, we show that segregation of the NK-mediated killing phenotype did not correlate with the NOD Raet1 haplotype and that the large alterations in NKG2D expression previously reported on NK cells expanded in vitro were not observed in primary, poly(I:C)-elicited NK cells in vivo. Additional studies indicate a complex genetic control of defective NOD NK cells including genes linked to the MHC and possibly those that are associated with an altered cytokine response to the TLR3-agonist poly(I:C).

Derivation of pluripotent epiblast stem cells from mammalian embryos

Although the first mouse embryonic stem (ES) cell lines were derived 25 years ago using feeder-layer-based blastocyst cultures, subsequent efforts to extend the approach to other mammals, including both laboratory and domestic species, have been relatively unsuccessful. The most notable exceptions were the derivation of non-human primate ES cell lines followed shortly thereafter by their derivation of human ES cells. Despite the apparent common origin and the similar pluripotency of mouse and human embryonic stem cells, recent studies have revealed that they use different signalling pathways to maintain their pluripotent status. Mouse ES cells depend on leukaemia inhibitory factor and bone morphogenetic protein, whereas their human counterparts rely on activin (INHBA)/nodal (NODAL) and fibroblast growth factor (FGF). Here we show that pluripotent stem cells can be derived from the late epiblast layer of post-implantation mouse and rat embryos using chemically defined, activin-containing culture medium that is sufficient for long-term maintenance of human embryonic stem cells. Our results demonstrate that activin/Nodal signalling has an evolutionarily conserved role in the derivation and the maintenance of pluripotency in these novel stem cells. Epiblast stem cells provide a valuable experimental system for determining whether distinctions between mouse and human embryonic stem cells reflect species differences or diverse temporal origins.

The derivation of highly germline-competent embryonic stem cells containing NOD-derived genome

It would be extremely advantageous to the analysis of disease mechanisms in the spontaneous mouse model of type 1 diabetes, the nonobese diabetic (NOD) strain, if genes in this strain could be modified in vivo using embryonic stem (ES) cells and homologous recombination. However, a NOD ES cell line with adequate germline transmission has not yet been reported. We report the development of highly germline-competent ES cell lines from the F1 hybrid of NOD and 129 for use in NOD gene targeting. Consequently, we developed ES cell lines derived from (NOD x 129)F1 x 129 backcross 1 mice, which were intercrossed to select for homozygosity of particular regions of NOD genome known to contain disease loci.

Assessment of the developmental potential of frozen-thawed mouse oocytes

Mouse oocytes were cryopreserved by a protocol shown previously to minimize damage to the zona pellucida and cytoskeletal system. After thawing, the incidence of fertilization did not differ from that in control groups of oocytes, and after fertilization, the ability of the fertilized frozen-thawed oocytes to develop to the blastocyst stage in vitro was only slightly less (77%) than that of the controls (87 and 89%). Transfer of frozen-thawed and fertilized oocytes after their culture to the blastocyst stage in vitro resulted in a lower implantation rate (46%) than for the controls (68-73%), but of the implanting embryos the same proportions in experimental and control groups survived to yield viable fetuses. In contrast, transfer after culture in vitro to the 2- to 4-cell stage resulted in similar implantation rates for control and frozen-thawed fertilized oocytes (70-84%), but the spontaneous abortion rate was higher for the embryos derived from frozen-thawed oocytes. Overall the cumulative survival rate for frozen oocytes transferred at the 2-cell stage (36%) was better than after transfer at the blastocyst stage (30%), but both were less than for the transfer at any stage of the control oocytes (47-55%). The cumulative survival of cryopreserved oocytes to viable fetuses was 30-40% less than that of the control oocytes. These results are compared with those from previous studies and the main remaining obstacles to completely successful cryopreservation are identified.

Adult phenotype in the mouse can be affected by epigenetic events in the early embryo

Major epigenetic modifications apparently occur during early development in the mouse. The factors that induce such modifications are complex and may involve the various components of a zygote. We have started to explore whether changes in the nucleocytoplasmic composition brought about by micromanipulation can induce phenotypic effects through epigenetic modifications. Nucleocytoplasmic hybrids were therefore prepared by transplanting a female pronucleus into a recipient egg from a different genotype. As a result, the maternal genome was of a different genetic background as compared with the egg cytoplasm. Specifically, experimental zygotes had cytoplasm from the inbred strain C57BL/6, a maternal genome from DBA/2, and a paternal genome from C57BL/6 (termed BDB hybrids). The mirror-image combination, termed DBD, was also made. The reconstituted zygotes were transferred to recipients and allowed to develop to term. Mice born from manipulated zygotes showed transcriptional repression and DNA methylation of major urinary protein genes in their liver, as well as growth deficiency resulting in reduced adult body weight. No altered phenotype was observed in controls in which the maternal pronucleus was simply transplanted back into another zygote of the same genetic background. These results clearly demonstrate phenotypic as well as molecular effects on DNA methylation and expression of at least one gene. Phenotype was therefore no longer predicted by genotype as a result of epigenetic modifications in experimental embryos. What precisely triggers the phenotypic and epigenetic changes is unknown, but presumably, nucleocytoplasmic interactions in hybrid zygotes may be partly responsible.

Developmental potential of parthenogenetic cells: role of genotype-specific modifiers

The developmental potential of parthenogenetic cells derived from different mouse strains was investigated by examining their distribution in various tissues of adult aggregation chimeras. Using GPI-1 allozymes as marker, no striking differences were observed between chimeras whose parthenogenetic cells were derived from activated oocytes isolated from females of different genetic backgrounds, (C57BL/6 × CBA/J) F1, CFLP, 129, and SWR. In all the combinations tested, parthenogenetic cells were consistently absent from skeletal muscle, but there were varying contributions to most other tissues. These results suggest that the maternal duplication of chromosomes containing imprinted gene(s) responsible for the systematic elimination of parthenogenetic cells from skeletal muscle, are not subject to a pronounced influence of genotype-specific modifiers. However, the contribution of parthenogenetic cells to the brain does appear to be influenced by strain background, since a marked improvement in the survival of CFLP, 129 and perhaps SWR parthenogenetic cells in chimeric brains was observed compared with F2 cells.

Methylation levels of maternal and paternal genomes during preimplantation development

The methylation status of three highly repeated sequences was studied in sperm, eggs and preimplantation embryos with different combinations of parental chromosomes. High levels of methylation of the IAP and MUP sequence families were found in sperm and in eggs, whereas the L1 repeat was found to be highly methylated in sperm but only about 42% methylated in eggs. To assess how the two parental genomes behaved during preimplantation development, normal, fertilised embryos were compared with parthenogenetic embryos where the chromosomes are exclusively of maternal origin. It was observed that the high levels of methylation at the IAP and MUP sequences were retained through early development, with the first signs of demethylation at the IAP sequences apparent on both parental chromosomes in the blastocyst. Methylation at the sperm-derived L1 sequences dropped to about the same level as that of the egg-derived sequences by the late 2-cell stage, both then remain at this intermediate level until around the time of cavitation when levels fell to about 10% in the blastocyst. High levels of DNA methylase were detected in germinal vesicle and metaphase II oocytes; these high levels were maintained in fertilised and parthenogenetic embryos through into the morula and then declined to be undetectable in the blastocyst. Our comparison of maternal and paternal genomes suggests that methylation levels at repeat sequences are remarkably similar at the time of fertilisation or, as in the case of the L1 sequences, they become so during the first few cell cycles. Hence, there do not appear to be global methylation differences between the genomes that are retained through preimplantation development.(ABSTRACT TRUNCATED AT 250 WORDS)

Imprinting by DNA methylation: from transgenes to endogenous gene sequences

A number of transgenes in the mouse show variation in methylation and expression phenotypes dependent on parental transmission. It appears that there exist at least two types of transgene imprinting; one is retained on an essentially homozygous background, while the other requires heterozygosity at some modifying loci in the genome and is observed as differences in phenotype in reciprocal crosses. For this type of imprinting to occur, the parental origin of the modifier locus itself is important, and parental asymmetry may involve specific interactions between egg cytoplasm and the chromosomes. Based on the identification of ‘methylation polymorphism’ in the mouse genome, we also show that endogenous gene sequences can undergo imprinting by DNA methylation.

Developmental consequences of imprinting of parental chromosomes by DNA methylation

Genomic imprinting by epigenetic modifications, such as DNA methylation, confers functional differences on parental chromosomes during development so that neither the male nor the female genome is by itself totipotential. We propose that maternal chromosomes are needed at the time when embryonic cells are totipotential or pluripotential, but paternal chromosomes are probably required for the proliferation of progenitor cells of differentiated tissues. Selective elimination or proliferation of embryonic cells may occur if there is an imbalance in the parental origin of some alleles. The inheritance of repressed and derepressed chromatin structures probably constitutes the initial germ-line-dependent 'imprints'. The subsequent modifications, such as changes in DNA methylation during early development, will be affected by the initial inheritance of epigenetic modifications and by the genotype-specific modifier genes. A significant number of transgene inserts are prone to reversible methylation imprinting so that paternally transmitted transgenes are undermethylated, whereas maternal transmission results in hypermethylation. Hence, allelic differences in epigenetic modifications can affect their potential for expression. The germ line evidently reverses the previously acquired epigenetic modifications before the introduction of new modifications. Errors in the reversal process could result in the transmission of epigenetic modifications to subsequent generation(s) with consequent cumulative phenotypic and grandparental effects.

Temporal and spatial selection against parthenogenetic cells during development of fetal chimeras

The fate of parthenogenetic cells was investigated during development of fetal and early postnatal chimeras. On day 13 of embryonic development, considerable contribution of parthenogenetic cells was observed in all tissues of chimeric embryos, although selection against parthenogenetic cells seemed to start before day 13. Between days 13 and 15 of development, parthenogenetic cells came under severe selective pressure, which was most striking in tongue. The disappearance of parthenogenetic cells from tongue coincided with the beginning of myoblast fusion in this tissue. Severe selection against parthenogenetic cells was also observed in pancreas and liver, although in the latter, parthenogenetic cells were eliminated later than in skeletal muscle or pancreas. In other tissues, parthenogenetic cells may persist and participate to a considerable extent throughout the gestation period and beyond, although a significant decrease was observed in all tissues. Parthenogenetic in equilibrium fertilized chimeras were significantly smaller than their non-chimeric littermates at all developmental stages. These results suggest that the absence of paternal chromosomes is largely incompatible with the maintenance of specific differentiated cell types. Furthermore, paternally derived genes seem to be involved in the regulation of proliferation of all cell types, as indicated by the drastic growth decceleration of parthenogenetic in equilibrium fertilized chimeras and the overall decrease of parthenogenetic cells during fetal development. Chromosomal imprinting may have a role in maintaining a balance between cell growth and differentiation during embryonic development. The major exception to the selective elimination of parthenogenetic cells appear to be the germ cells; viable offspring derived from parthenogenetic oocytes were detected, sometimes at a high frequency in litters of female parthenogenetic in equilibrium fertilized chimeras.

Influence of chromosomal determinants on development of androgenetic and parthenogenetic cells

We have examined the role of germline-specific chromosomal determinants of development in the mouse. Studies were carried out using aggregation chimaeras between androgenetic----fertilized embryos and compared with similar parthenogenetic----fertilized chimaeras. Several adult chimaeras were found with parthenogenetic cells but none were found with androgenetic cells. Analysis of chimaeras at mid-gestation showed that parthenogenetic cells were detected in the embryo and yolk sac but that androgenetic cells were found only in the trophoblast and yolk sac and not in the embryo. The contribution of parthenogenetic cells to the embryo and yolk sac was increased by aggregating 2-cell parthenogenetic and 4-cell fertilized embryos but the contribution of parthenogenetic cells in extraembryonic tissues remained negligible even after aggregation of 4-cell parthenogenetic and 2-cell fertilized embryos. Furthermore, parthenogenetic cells were primarily found in the yolk sac mesoderm and not in the yolk sac endoderm. These results suggest that maternal chromosomes in parthenogenetic cells permit their participation in the primitive ectoderm lineage but these cells are presumably eliminated by selective pressure or autonomous cell lethality from the primitive endoderm and trophectoderm lineages. Conversely paternal chromosomes in androgenetic cells confer opposite properties since the embryonic cells can be detected in the trophoblast and the yolk sac but not in the embryos, presumably because they are eliminated from the primitive ectoderm lineage. The spatial distribution of cells with different parental chromosomes may occur partly because of differential expression of some genes, such as proto-oncogenes, and partly due to their ability to respond to a variety of diffusible growth factors.

Nuclear cytoplasmic interactions following nuclear transplantation in mouse embryos

We have investigated the development of reconstituted embryos in which enucleated 1- or 2-cell embryos received various advanced nuclei. Enucleated 1-cells developed to the blastocyst stage only when an early 2-cell donor nucleus was transferred but very rarely if the donor nucleus was derived from a late 2-cell, early 4-cell or mid 8-cell embryo. Although an 8-cell nucleus could only support development of an enucleated zygote to the 2-cell stage, it did express the hsp 68/70 X 10(3) Mr proteins that are characteristic of the first embryonic gene activity. These polypeptides were absent in enucleated zygotes that did not receive a donor nucleus. Moreover, an 8-cell nucleus transferred to an enucleated late 2-cell blastomere could also support preimplantation development provided that the nuclear:cytoplasmic ratio was maintained as in intact 2-cell blastomeres. 8-cell nuclei transferred to zygotes that retained at least one pronucleus were able to support development to the blastocyst stage provided that the pronucleus was both fully transcriptionally active and present beyond the late 1-cell stage. This study suggests an active and continued helper role of the resident pronucleus for the participation by an 8-cell nucleus in reconstituted eggs.

The effect of inhibiting DNA replication in the one-cell mouse embryo

Inhibition of DNA replication in the 1-cell mouse embryo causes arrest in interphase. In such arrested cells the pattern of protein synthesis progresses as normal through the first cell cycle but does not show the characteristic changes that are associated with mitosis. Synthesis does occur of the 68/70 kd heat shock-like proteins, on the earliest detectable embryonic transcripts synthesised shortly after first cleavage in control embryos. However, the quantitative change in the pattern of protein synthesis, indicative of the major activation of the embryonic genome that normally occurs during G2 of the second cell cycle, is prevented.

Parthenogenesis and cytoskeletal organization in ageing mouse eggs

The cytoskeletal organization of the mouse egg changes during ageing in vivo and in vitro. The earliest change observed is the disappearance of the microfilament-rich area overlying the meiotic spindle. This is followed by the migration of the spindle towards the centre of the egg. Finally the spindle breaks down and the chromosomes are no longer organized on a metaphase plate. This spindle disruption may result from changes in the microtubule nucleating material found at the spindle poles and from an increase in the critical concentration for tubulin polymerization. It is possible to correlate the changes in the cytoskeletal organization of the egg occurring during ageing with the different types of parthenogenetic embryos obtained after ethanol activation. These observations strengthen the hypothesis that the actin-rich cortical area that overlies the meiotic spindle forms a domain to which the meiotic cleavage furrow is restricted and provides some insights into the mechanisms by which different types of parthenogenetic embryos are generated.

Parthenogenesis and cytoskeletal organization in ageing mouse eggs

The cytoskeletal organization of the mouse egg changes during ageing in vivo and in vitro. The earliest change observed is the disappearance of the microfilament-rich area overlying the meiotic spindle. This is followed by the migration of the spindle towards the centre of the egg. Finally the spindle breaks down and the chromosomes are no longer organized on a metaphase plate. This spindle disruption may result from changes in the microtubule nucleating material found at the spindle poles and from an increase in the critical concentration for tubulin polymerization. It is possible to correlate the changes in the cytoskeletal organization of the egg occurring during ageing with the different types of parthenogenetic embryos obtained after ethanol activation. These observations strengthen the hypothesis that the actin-rich cortical area that overlies the meiotic spindle forms a domain to which the meiotic cleavage furrow is restricted and provides some insights into the mechanisms by which different types of parthenogenetic embryos are generated.

A set of proteins showing cell cycle dependent modification in the early mouse embryo

The pattern of protein synthesis in the mouse egg shows several changes at fertilization and during first mitosis. Three groups of newly synthesized proteins, with molecular weights of about 30,000, 35,000, and 46,000, show variations in mobility on one- and two-dimensional gels that correlate with the cell cycle. Each group is composed of a polypeptide that is synthesized in unmodified form during interphase but is modified reversibly during meiosis or mitosis, by a process involving phosphorylation. Although these proteins cease to be synthesized during the second cell cycle, those made earlier persist and continue to show the same modifications during the next cell cycle. Like other eggs, fertilized mouse eggs show a requirement for protein synthesis in order to enter mitosis.

Cytoskeletal dynamics in the mouse egg

The distribution and roles of the microtubule and microfilament networks in the mouse egg following fertilization are described. The role of the chromosomes in the control of the egg cytoskeleton organization is discussed and a model for polar body formation proposed. Finally we described the changes occurring in the pattern of proteins synthesized during this period, these being discussed in relation to cell cycle events and to change in cytoskeleton organization.

Non-spindle microtubule organizing centers in metaphase II-arrested mouse oocytes

A human autoantiserum (5051) directed against pericentriolar material (PCM) was used to study the distribution of microtubule-organizing centers (MTOCs) in the oocyte and during the first cell cycle of mouse development. In oocytes, the PCM was found not only at the poles of the barrel-shaped metaphase II spindle but also at many discrete loci around the cytoplasm near the cell cortex. The spindle poles were also composed of several PCM foci. In metaphase-arrested eggs only the PCM foci located near the chromosomes acted as MTOCs. However, after reduction of the critical concentration for tubulin polymerization by taxol, the cytoplasmic PCM foci were also found to be associated with nucleation of microtubules. After fertilization the cortical PCM foci remained in a peripheral position until the end of the S phase, when they appeared to migrate centrally towards the pronuclei. At prometaphase of the first mitotic division, numerous MTOCs were found around the two sets of chromosomes; these MTOCs then aligned to form two bands on either side of the metaphase plate of the first mitosis.

Sequence and regulation of morphological and molecular events during the first cell cycle of mouse embryogenesis

Mouse oocytes were fertilized in vitro and the precise timing and sequence of morphological and molecular events occurring during the first cell cycle were investigated. The timing of development through the first cell cycle was found to be initiated by an event associated with sperm penetration rather than with germinal vesicle breakdown. DNA replication is initiated randomly in either pronucleus of a given egg, beginning approximately 11 h post insemination (hpi), and S phase lasting 6-7 h in both. Careful study of polypeptide synthetic profiles revealed three classes of changes in polypeptide synthesis during the first few hours of development: fertilization-independent, fertilization-accelerated, and fertilization-dependent. Pulse-chase experiments and in vitro translation of extracted mRNA showed that the changes in polypeptide synthetic profile result from differential mRNA activation, differential polypeptide turnover and post-translational modifications. These results support the notion that following ovulation, development is controlled at two levels. An endogenous (oocyte) programme, set in train by the terminal events of oocyte maturation, may regulate the 'housekeeping' functions of the egg, while sperm penetration activates a further endogenous (fertilization) programme, which may serve to initiate subsequent embryogenesis.

Sequence and regulation of morphological and molecular events during the first cell cycle of mouse embryogenesis

Mouse oocytes were fertilized in vitro and the precise timing and sequence of morphological and molecular events occurring during the first cell cycle were investigated. The timing of development through the first cell cycle was found to be initiated by an event associated with sperm penetration rather than with germinal vesicle breakdown. DNA replication is initiated randomly in either pronucleus of a given egg, beginning approximately 11 h post insemination (hpi), and S phase lasting 6–7 h in both. Careful study of polypeptide synthetic profiles revealed three classes of changes in polypeptide synthesis during the first few hours of development: fertilization-independent, fertilization-accelerated, and fertilization-dependent. Pulse-chase experiments and in vitro translation of extracted mRNA showed that the changes in polypeptide synthetic profile result from differential mRNA activation, differential polypeptide turnover and post-translational modifications. These results support the notion that following ovulation, development is controlled at two levels. An endogenous (oocyte) programme, set in train by the terminal events of oocyte maturation, may regulate the ‘housekeeping’ functions of the egg, while sperm penetration activates a further endogenous (fertilization) programme, which may serve to initiate subsequent embryogenesis.