Response to discussants of "survival models and health sequences".

Survival studies often generate not only a survival time for each patient but also a sequence of health measurements at annual or semi-annual check-ups while the patient remains alive. Such a sequence of random length accompanied by a survival time is called a survival process. Robust health is ordinarily associated with longer survival, so the two parts of a survival process cannot be assumed independent. This paper is concerned with a general technique-reverse alignment-for constructing statistical models for survival processes, here termed revival models. A revival model is a regression model in the sense that it incorporates covariate and treatment effects into both the distribution of survival times and the joint distribution of health outcomes. The revival model also determines a conditional survival distribution given the observed history, which describes how the subsequent survival distribution is determined by the observed progression of health outcomes.

Survival models and health sequences.

Survival studies often generate not only a survival time for each patient but also a sequence of health measurements at annual or semi-annual check-ups while the patient remains alive. Such a sequence of random length accompanied by a survival time is called a survival process. Robust health is ordinarily associated with longer survival, so the two parts of a survival process cannot be assumed independent. This paper is concerned with a general technique-reverse alignment-for constructing statistical models for survival processes, here termed revival models. A revival model is a regression model in the sense that it incorporates covariate and treatment effects into both the distribution of survival times and the joint distribution of health outcomes. The revival model also determines a conditional survival distribution given the observed history, which describes how the subsequent survival distribution is determined by the observed progression of health outcomes.

Effects of an Extended Cage-change Interval on Ammonia Levels and Reproduction in Mongolian Gerbils (Meriones unguiculatus).

Prompted by the cage cleanliness of Mongolian gerbils (Meriones unguiculatus), we evaluated a prolonged cage-change interval. We compared the effects of a 2-wk and 6-wk cage-change schedule on ammonia levels, temperature, humidity, and reproductive performance in breeding pairs housed in IVC. We hypothesized that ammonia levels would remain below our threshold for cage changing and that reproductive performance would not be affected. Although ammonia levels increased over time, they remained low (less than 5 ppm) over the 6-wk period. In addition, the 6-wk cage-change interval did not significantly influence reproductive parameters, such as average pup weaning weight, number of litters, and number of pups per litter. We conclude that an extended cage-change interval (6-wk) can be used for gerbils without significant increases in intracage ammonia levels or effects on reproduction.

Exchangeable Markov survival processes and weak continuity of predictive distributions.

We study exchangeable, Markov survival processes - stochastic processes giving rise to infinitely exchangeable non-negative sequences (T 1, T 2, …). We show how these are determined by their characteristic index { ζ n } n = 1 ∞ . We identify the harmonic process as the family of exchangeable, Markov survival processes that compose the natural set of statistical models for time-to-event data. In particular, this two-dimensional family comprises the set of exchangeable, Markov survival processes with weakly continuous predictive distributions. The harmonic process is easy to generate sequentially, and a simple expression exists for both the joint probability distribution and multivariate survivor function. We show a close connection with the Kaplan-Meier estimator of the survival distribution. Embedded within the process is an infinitely exchangeable ordered partition. Aspects of the process, such as the distribution of the number of blocks, are investigated.

Deriving Coarse-Grained Charges from All-Atom Systems: An Analytic Solution.

An analytic method to assign optimal coarse-grained charges based on electrostatic potential matching is presented. This solution is the infinite size and density limit of grid-integration charge-fitting and is computationally more efficient by several orders of magnitude. The solution is also minimized with respect to coarse-grained positions which proves to be an extremely important step in reproducing the all-atom electrostatic potential. The joint optimal-charge optimal-position coarse-graining procedure is applied to a number of aggregating proteins using single-site per amino acid resolution. These models provide a good estimate of both the vacuum and Debye-Hückel screened all-atom electrostatic potentials in the vicinity and in the far-field of the protein. Additionally, these coarse-grained models are shown to approximate the all-atom dimerization electrostatic potential energy of 10 aggregating proteins with good accuracy.

Evidence that Notch and Delta expressions have a role in dermal condensate aggregation during wool follicle initiation.

Notch pathway genes have been implicated in the commitment of mesenchymal cells to a wool follicle cell fate. Notch1 and Delta1 transcripts were quantified in fetal skin of fine-woolled (Merino) and strong-woolled (Tukidale) sheep at two time points: either preceding (d56) or during (d70) the first wave of follicle initiation. DIG-labelled probes for both transcripts were localised in the epithelium, some mesenchymal cells, and in the dermal condensates of primordia. The possibility that condensates selectively incorporated Delta1-labelled mesenchymal cells is considered. The involvement of Notch1 in condensate formation was also explored in cultured fetal skin explants and whisker papilla cells using DAPT to block Notch signalling. In its presence, follicle initiation in skin explants was reduced, and the propensity for cultured papilla cells to aggregate was abolished. Results suggest that Notch1 activation is a prerequisite for mesenchymal aggregation. It is speculated that Delta interactions contribute to condensate formation, in vivo.

Establishment of an in vitro system representing the chicken gut-associated lymphoid tissue.

The bursa of Fabricius is critical for B cell development and differentiation in chick embryos. This study describes the production in vitro, from dissociated cell suspensions, of cellular agglomerates with functional similarities to the chicken bursa. Co-cultivation of epithelial and lymphoid cells obtained from embryos at the appropriate developmental stage regularly led to agglomerate formation within 48 hours. These agglomerates resembled bursal tissue in having lymphoid clusters overlaid by well organized epithelium. Whereas lymphocytes within agglomerates were predominantly Bu-1a(+), a majority of those emigrating onto the supporting membrane were Bu-1a(-) and IgM(+). Both agglomerates and emigrant cells expressed activation-induced deaminase with levels increasing after 24 hours. Emigrating cells were actively proliferating at a rate in excess of both the starting cell population and the population of cells remaining in agglomerates. The potential usefulness of this system for investigating the response of bursal tissue to avian Newcastle disease virus (strain AF2240) was examined.

B-cell development: one problem, multiple solutions.

Interspecies variations in the processes of B-cell development and repertoire generation contrast with the greater consistency of T-cell development. B-cell development in mice and humans, with postnatal B-cell generation of new repertoire in the bone marrow throughout life, is regarded as the 'standard' pattern. In contrast, accounts of B cells in birds, sheep, cattle, rabbits and pigs (the 'other' species) describe cessation of gene diversification in the perinatal period, with the gut-associated lymphoid tissue (GALT) functioning as the primary lymphoid organ thereafter. It has become customary to regard the developmental pathways of T and B cells within any individual species as being as dissimilar as the functions of the two mature cell types. Reinterpretation of B-cell development patterns in different species is overdue in response to two types of reports. The first of these describe T-B 'crossover', specifically the intrathymic production of B cells and the extrathymic production of T cells. The second attests to the extent of sharing of B-cell developmental features across the two groups of species. We propose that, as is a feature of other haematopoietic cells, a menu of alternative B- and T-cell pathways has been retained and shared across species. A single pathway usually predominates in any species, masking alternatives. The observed predominance of any pathway is determined by factors such as placental permeability, extent of maturation of the immune system by birth and the feasibility of direct experimental intervention in development.

Detection and quantification of IgM(+) lymphocytes in fetal lamb spleen, liver and lymph nodes by flow cytometry.

The first stage in Peyer's patch development in the fetal lamb is characterized by the colonization of the rudimentary Peyer's patches by precursor cells expressing the IgM surface receptor. In the fetal lamb, the spleen has been implicated as the source of gene-rearranged IgM(+) B lymphocytes. This study was intended to quantitate IgM(+) lymphocytes in the spleen, lymph nodes and liver of fetal lambs at various gestational ages between 63 and 110 days using flow cytometry. Flow-cytometric analysis revealed that IgM(+) lymphocytes were rare in the liver being consistently less than 1% at every gestational age examined. IgM(+) lymphocytes were detected in the spleen (mean 9.18%) and prescapular lymph nodes (mean 11.89%) as early as 63 days. In both spleen and lymph nodes, the highest representation of IgM(+) lymphocytes occurred between 70 and 86 days gestation. The highest mean percentage of IgM(+) lymphocytes was observed in the spleen (22.63%) and lymph nodes (17.02%) at 75 days gestation. From 98 days onwards, B-lymphocyte density gradually decreased in both spleen and prescapular lymph nodes. This study indicates that substantial populations of IgM(+) lymphocytes were present in both the spleen and prescapular lymph nodes from 70 days gestation and implies that both of these locations could be potential sources for the normal colonization of the ileal Peyer's patches.

Transient transmission of porcine endogenous retrovirus to fetal lambs after pig islet tissue xenotransplantation.

Evidence for the in vivo transmission of porcine endogenous retrovirus (PERV) from porcine xenografts to various recipient animals has been inconsistent. To characterize the contribution of the host immune system to the potential for PERV transmission from pig islet tissue xenografts to host tissues, we examined two immunoincompetent animal models, thymectomizsed fetal lambs and NODscid mice. Pig proislets were grafted into fetal lambs or adult NODscid mice. Conventional, nested and real-time PCR/RT-PCR tests were used to search for PERV and pig cell-specific sequences (porcine mitochondrial cytochrome oxidase II (COII) or mitochondrial ribosomal 12S) in pig proislets, host liver and spleen at 5-84 days (lambs) or 96 days (mice) after transplantation. Xenografts were harvested at the same time points. The copy number of PERV sequences and host cell-specific nuclear (palmitoylcarnitine transferase) sequences was assessed by real-time PCR to estimate the proportion of PERV-infected host cells. Pig proislets were shown to be PERV+ve by PCR and immunohistochemistry (PERV B env protein p15E). PERV transmission (PERV A, B or C DNA in the absence of porcine COII or 12S sequences) was detected by nested PCR and real-time PCR in 4/12 fetal lamb liver samples 5-23 days after transplantation; the maximum copy number of PERV B env sequences was found at day 5 (700 copies/1 x 10(6) lamb cells). A total of 4/12 fetal lambs demonstrated both PERV and 12S porcine sequences in liver samples (days 5-84) by real-time PCR, suggesting that pig cells had migrated to those tissues and established microchimerism; nested PCR showed evidence for microchimerism (porcine COII sequences alone) in 2/12 lambs (day 5). The incidence of PERV transmission and frequency of microchimerism was similar in host spleen analysed by real-time PCR. Histological examination showed complete xenograft rejection by 23 days after transplantation to fetal lambs. In contrast, pig proislet xenografts survived long term (> or =day 96) in NODscid mice but no PERV transmission was found. Both nested and real-time PCR assays revealed that 2/3 mice had become microchimeric. Long-term expression of PERV A, B and C as well as porcine 12S or COII RNAs was found at the graft site (day 96) only, indicating that PERV transcription and possibly replication, continued in the donor pig islet tissue after transplantation. Overall, detection of PERV transmission and microchimerism was limited by the sensitivity of the PCR assay and the primers chosen. The absence of stable PERV transmission and microchimerism in fetal lambs and the rejection of pig proislet xenografts correlated in time with the establishment of host immunocompetence. We therefore suggest that the frequent failure to identify PERV transmission late after transplantation could be due to the immunological destruction of PERV-infected host cells. Recipient NODscid mice demonstrated long-term microchimerism and intragraft PERV expression, which was consistent with their stable immunoincompetence.

Therapeutic potential of stem cells in perinatal medicine.

Increasing evidence suggests that stem cells have tremendous potential to facilitate repair of damaged tissue and to exert protective influences that limit the extent of damage. Their inherent capacity to respond to signals generated by damaged tissue, migrate to these regions and either replace dead tissue or deliver protection by secretion of specific growth hormones and protective factors, suggests that they might have unrivalled therapeutic potential in perinatal medicine. A further potential of stem cells is their use in gene repair strategies for genetic disorders; an application which is exceedingly interesting from a perinatal perspective. Because of the relatively small size of infants and their capacity for future growth, stem cell therapy could be more successful in newborns than in older children or adults. In practical terms, the placenta, with its large reservoir of fetal blood, offers the ideal source of autologous stem cells. This affords the opportunity for stem cells to be collected and used, either directly ex vivo or after in vitro modulation, both for disorders in the neonatal period and for those arising later in life. The organs most affected from tissue damage in the neonatal period are the brain and the lung. So far, the most promising application of stem cells might be in the treatment of neurological injury. In this review we discuss recent research findings with adult stem cell therapy and their potential use in perinatal medicine. Furthermore, specific animal models suitable to explore the patho-physiological mechanisms of stem cell transplantation after neurological injury will be discussed. This review gives an overview of basic science findings and their possible role for clinical application with regards to the therapeutic potential of stem cells in perinatal medicine. Medline was searched for journal selection in peer-reviewed journals with high impact scores, which were relevant to this topic. All articles were in English and the search was not limited by publication year. However, the oldest publication was dated 1988 (reference 1).

Intestinal exposure to a parasite antigen in utero depresses cellular and cytokine responses of the mucosal immune system.

The response of the mucosal immune system of 4-6-week old lambs to viable Trichostrongylus colubriformis larvae was compared in two groups of animals, one exposed to T. colubriformis antigen and the other to saline while in utero. Exposure to larval antigen two-thirds of the way through gestation resulted in significant reduction in the frequency of jejunal goblet cells and of ileal eosinophils, CD 1b(+) antigen-presenting cells and CD4(+), CD5(+) and CD8(+) cells. However, it resulted in a significant increase in the jejunal CD8(+) response to postnatal challenge. The expression of the cytokines TNF-alpha and IL-1 beta in the ileum, and of jejunal NSE, was significantly reduced by in utero exposure, whereas those of jejunal TNF-alpha and ileal TGF-beta were increased. The observed changes in cellular and cytokine responses to challenge with viable larvae, in those lambs previously exposed in utero, indicated that the intestinal mucosal immune system remains susceptible to down-regulation until considerably later in foetal development than is the case for other components of the immune system.

Development of B cells in the gut-associated lymphoid tissue of mid-gestational fetal lambs.

Intestinal mucosal immune system development was investigated in fetal lambs from 61 to 110 days (term, 150 days). Fetal small intestine was examined at 400 mm intervals for the presence of IgM(+) cells. The first phase of B cell development was characterised by increase in B cell density throughout the jejunum from 65 days. Increase in density was greatest in the proximal jejunum and declined progressively approaching the ileum. The second phase entailed a decrease in jejunal B cell concentration, evident from 90 days. The average number of cells per field diminished, by 110 days, to a 10th that at 90 days. Failure of B cell increase to match a five-fold intestinal lengthening may have contributed to this. Overlapping the two phases of jejunal B cell development was a third phase of major expansion of B cell density in the terminal ileum.

A new model of sheep Ig diversification: shifting the emphasis toward combinatorial mechanisms and away from hypermutation.

The current model of Ig repertoire development in sheep focuses on the rearrangement of a small number (approximately 20) of Vlambda gene segments. It is believed that this limited combinatorial repertoire is then further diversified through postrearrangement somatic hypermutation. This process has been reported to introduce as many as 110 mutations/1000 nucleotides. In contrast, our data have that indicated somatic hypermutation may diversify the preimmune repertoire to a much lesser extent. We have identified 64 new Vlambda gene segments within the rearranged Ig repertoire. As a result, many of the unique nucleotide patterns thought to be the product of somatic hypermutation are actually hard-coded within the germline. We suggest that combinatorial rearrangement makes a much larger contribution, and somatic hypermutation makes a much smaller contribution to the generation of diversity within the sheep Ig repertoire than is currently acknowledged.