Longitudinal analytical approaches to genetic data

Background

Longitudinal phenotypic data provides a rich potential resource for genetic studies which may allow for greater understanding of variants and their covariates over time. Herein, we review 3 longitudinal analytical approaches from the Genetic Analysis Workshop 19 (GAW19). These contributions investigated both genome-wide association (GWA) and whole genome sequence (WGS) data from odd numbered chromosomes on up to 4 time points for blood pressure–related phenotypes. The statistical models used included generalized estimating equations (GEEs), latent class growth modeling (LCGM), linear mixed-effect (LME), and variance components (VC). The goal of these analyses was to test statistical approaches that use repeat measurements to increase genetic signal for variant identification.

Results

Two analytical methods were applied to the GAW19: GWA using real phenotypic data, and one approach to WGS using 200 simulated replicates. The first GWA approach applied a GEE-based model to identify gene-based associations with 4 derived hypertension phenotypes. This GEE model identified 1 significant locus, GRM7, which passed multiple test corrections for 2 hypertension-derived traits. The second GWA approach employed the LME to estimate genetic associations with systolic blood pressure (SBP) change trajectories identified using LCGM. This LCGM method identified 5 SBP trajectories and association analyses identified a genome-wide significant locus, near ATOX1 (p?=?1.0E^?8). Finally, a third VC-based model using WGS and simulated SBP phenotypes that constrained the β coefficient for a genetic variant across each time point was calculated and compared to an unconstrained approach. This constrained VC approach demonstrated increased power for WGS variants of moderate effect, but when larger genetic effects were present, averaging across time points was as effective.

Conclusion

In this paper, we summarize 3 GAW19 contributions applying novel statistical methods and testing previously proposed techniques under alternative conditions for longitudinal genetic association. We conclude that these approaches when appropriately applied have the potential to: (a) increase statistical power; (b) decrease trait heterogeneity and standard error; (c) decrease computational burden in WGS; and (d) have the potential to identify genetic variants influencing subphenotypes important for understanding disease progression.

     

LKB1 is required for hepatic bile acid transport and canalicular membrane integrity in mice

LKB1 is a 'master' protein kinase implicated in the regulation of metabolism, cell proliferation, cell polarity and tumorigenesis. However, the long-term role of LKB1 in hepatic function is unknown. In the present study, it is shown that hepatic LKB1 plays a key role in liver cellular architecture and metabolism. We report that liver-specific deletion of LKB1 in mice leads to defective canaliculi and bile duct formation, causing impaired bile acid clearance and subsequent accumulation of bile acids in serum and liver. Concomitant with this, it was found that the majority of BSEP (bile salt export pump) was retained in intracellular pools rather than localized to the canalicular membrane in hepatocytes from LLKB1KO (liver-specific Lkb1-knockout) mice. Together, these changes resulted in toxic accumulation of bile salts, reduced liver function and failure to thrive. Additionally, circulating LDL (low-density lipoprotein)-cholesterol and non-esterified cholesterol levels were increased in LLKB1KO mice with an associated alteration in red blood cell morphology and development of hyperbilirubinaemia. These results indicate that LKB1 plays a critical role in bile acid homoeostasis and that lack of LKB1 in the liver results in cholestasis. These findings indicate a novel key role for LKB1 in the development of hepatic morphology and membrane targeting of canalicular proteins.     

Functional Evaluation of Biological Neurotoxins in Networked Cultures of Stem Cell-derived Central Nervous System Neurons

Therapeutic and mechanistic studies of the presynaptically targeted clostridial neurotoxins (CNTs) have been limited by the need for a scalable, cell-based model that produces functioning synapses and undergoes physiological responses to intoxication. Here we describe a simple and robust method to efficiently differentiate murine embryonic stem cells (ESCs) into defined lineages of synaptically active, networked neurons. Following an 8 day differentiation protocol, mouse embryonic stem cell-derived neurons (ESNs) rapidly express and compartmentalize neurotypic proteins, form neuronal morphologies and develop intrinsic electrical responses. By 18 days after differentiation (DIV 18), ESNs exhibit active glutamatergic and γ-aminobutyric acid (GABA)ergic synapses and emergent network behaviors characterized by an excitatory:inhibitory balance. To determine whether intoxication with CNTs functionally antagonizes synaptic neurotransmission, thereby replicating the in vivo pathophysiology that is responsible for clinical manifestations of botulism or tetanus, whole-cell patch clamp electrophysiology was used to quantify spontaneous miniature excitatory post-synaptic currents (mEPSCs) in ESNs exposed to tetanus neurotoxin (TeNT) or botulinum neurotoxin (BoNT) serotypes /A-/G. In all cases, ESNs exhibited near-complete loss of synaptic activity within 20 hr. Intoxicated neurons remained viable, as demonstrated by unchanged resting membrane potentials and intrinsic electrical responses. To further characterize the sensitivity of this approach, dose-dependent effects of intoxication on synaptic activity were measured 20 hr after addition of BoNT/A. Intoxication with 0.005 pM BoNT/A resulted in a significant decrement in mEPSCs, with a median inhibitory concentration (IC₅₀) of 0.013 pM. Comparisons of median doses indicate that functional measurements of synaptic inhibition are faster, more specific and more sensitive than SNARE cleavage assays or the mouse lethality assay. These data validate the use of synaptically coupled, stem cell-derived neurons for the highly specific and sensitive detection of CNTs.     

Embryonic Lethality in Homozygous Human Her-2 Transgenic Mice Due to Disruption of the Pds5b Gene

The development of antigen-targeted therapeutics is dependent on the preferential expression of tumor-associated antigens (TAA) at targetable levels on the tumor. Tumor-associated antigens can be generated de novo or can arise from altered expression of normal basal proteins, such as the up-regulation of human epidermal growth factor receptor 2 (Her2/ErbB2). To properly assess the development of Her2 therapeutics in an immune tolerant model, we previously generated a transgenic mouse model in which expression of the human Her2 protein was present in both the brain and mammary tissue. This mouse model has facilitated the development of Her2 targeted therapies in a clinically relevant and suitable model. While heterozygous Her2^+/- mice appear to develop in a similar manner to wild type mice (Her2^-/-), it has proven difficult to generate homozygous Her2^+/+ mice, potentially due to embryonic lethality. In this study, we performed whole genome sequencing to determine if the integration site of the Her2 transgene was responsible for this lethality. Indeed, we report that the Her2 transgene had integrated into the Pds5b (precocious dissociation of sisters) gene on chromosome 5, as a 162 copy concatemer. Furthermore, our findings demonstrate that Her2^+/+ mice, similar to Pds5b^-/- mice, are embryonic lethal and confirm the necessity for Pds5b in embryonic development. This study confirms the value of whole genome sequencing in determining the integration site of transgenes to gain insight into associated phenotypes.     

Colonization of Lutzomyia verrucarum and Lutzomyia longipalpis Sand Flies (Diptera: Psychodidae) by Bartonella bacilliformis,the Etiologic Agent of Carrión’s Disease

Bartonella bacilliformis is a pathogenic bacterium transmitted to humans presumably by bites of phlebotomine sand flies, infection with which results in a bi-phasic syndrome termed Carrión’s disease. After constructing a low-passage GFP-labeled strain of B. bacilliformis, we artificially infected Lutzomyia verrucarum and L. longipalpis populations, and subsequently monitored colonization of sand flies by fluorescence microscopy. Initially, colonization of the two fly species was indistinguishable, with bacteria exhibiting a high degree of motility, yet still confined to the abdominal midgut. After 48h, B. bacilliformis transitioned from bacillus-shape to a non-motile, small coccoid form and appeared to be digested along with the blood meal in both fly species. Differences in colonization patterns became evident at 72h when B. bacilliformis was observed at relatively high density outside the peritrophic membrane in the lumen of the midgut in L. verrucarum, but colonization of L. longipalpis was limited to the blood meal within the intra-peritrophic space of the abdominal midgut, and the majority of bacteria were digested along with the blood meal by day 7. The viability of B. bacilliformis in L. longipalpis was assessed by artificially infecting, homogenizing, and plating for determination of colony-forming units in individual flies over a 13-d time course. Bacteria remained viable at relatively high density for approximately seven days, suggesting that L. longipalpis could potentially serve as a vector. The capacity of L. longipalpis to transmit viable B. bacilliformis from infected to uninfected meals was analyzed via interrupted feeds. No viable bacteria were retrieved from uninfected blood meals in these experiments. This study provides significant information toward understanding colonization of sand flies by B. bacilliformis and also demonstrates the utility of L. longipalpis as a user-friendly, live-vector model system for studying this severely neglected tropical disease.     

Structure and sequence of the human α- -iduronidase gene

In humans, a deficiency of the lysosomal hydrolase α- -iduronidase (IDUA; EC 3.2.1.76) results in the lysosomal storage of the glycosaminoglycans heparan sulfate and dermatan sulfate, thereby causing the lysosomal storage disorder mucopolysaccharidosis type I. The gene for IDUA is split into 14 exons spanning approximately 19 kb. We report the sequence of two noncontiguous segments of the IDUA gene, one 1.8-kb segment containing exons 1 and 2 and surrounding sequences and a second segment of 4.5 kb containing the last 12 exons. The potential promoter for IDUA has only GC box type consensus sequences consistent with a housekeeping promoter and is bounded by an Alu repeat sequence. The first two exons of IDUA are separated by an intron of 566 bp, then there is a large intron of approximately 13 kb, and the last 12 exons are clustered within 4.5 kb. No consensus polyadenylation signal was found in the 3′ untranslated region, although two variant polyadenylation signals are proposed.     

Field estimates of reproductive success in a model insect: behavioural surrogates are poor predictors of fitness

Understanding, and therefore measuring, factors that determine fitness is a central problem in evolutionary biology. We studied a natural population of Coenagrion puella (Odonata: Zygoptera) over two entire breeding seasons, with over a thousand individuals uniquely marked and genotyped, and all mating events at the rendezvous site recorded. Using a parentage analysis, fitness of individuals in the first generation was quantified as the numbers of offspring that survived to maturity. Although mating behaviour can be predicted by environmental and demographical variables, the numbers of mature offspring produced (fitness) cannot, and crucially, are poorly correlated with behavioural observations of mating. While fitness of both sexes was positively related to mating behaviour and to female's ectoparasite burden, these behavioural observations explained little more variance in offspring production than environmental and demographical variables. Thus, we demonstrate that behavioural measures of reproductive success are not necessarily reliable estimates of fitness in natural populations.     

Crossmodal session rating of perceived exertion response at low and moderate intensities

Session rating of perceived exertion (SRPE) permits global effort estimations after an exercise bout and has shown promise for evaluating training load. However, factors mediating SRPE are not well understood. The purpose of this study was to compare SRPE between cycling and treadmill exercise at low and moderate intensities. In a counterbalanced order, male subjects (n = 7) completed a VO2max trial on a cycle ergometer and a motor-driven treadmill. Then, participants completed trials at 50 and 75% mode-specific VO2max on a cycle ergometer (BK75, BK50) and a treadmill (TM75, TM50) to achieve ～ 400-kcal energy expenditure per trial. Acute RPE (i.e., during exercise) at 5 minutes, midway, and test termination were recorded with SRPE (20-minutes postexercise) expressed as overall (SRPEO), legs (SRPEL), and breathing also recorded were heart rate (HR) and change in rectal temperature (ΔTrec). Significance was accepted at p ≤ 0.05. Repeated-measures analysis of variance revealed significantly greater SRPE for higher intensities within each mode. Crossmodal comparisons also show a higher SRPE at moderate (75% VO2max) intensities [SRPEO] = BK75: 7.6 ± 1.0, TM75: 6.9 ± 1.3) vs. lower (50% VO2max) intensities (BK50: 4.6 ± 1.4, TM50: 4.6 ± 1.1). Within modes, SRPE corresponded well with ΔTrec and HR. Acute RPE was linked with intensity and drifted upward across time. Results indicated that overall and differentiated SRPEs are magnified with exercise intensity with the corresponding disruption in internal environment potentially mediating subjective responses. From a practical application standpoint, SRPE provides a subjective assessment for immediate evaluation of daily training. Results indicate that, when using SRPE to monitor training, consideration should be given to responses across differing exercise modes.     

Large-scale analysis of yeast filamentous growth by systematic gene disruption and overexpression

Under certain conditions of nutrient stress, the budding yeast Saccharomyces cerevisiae initiates a striking developmental transition to a filamentous form of growth, resembling developmental transitions required for virulence in closely related pathogenic fungi. In yeast, filamentous growth involves known mitogen-activated protein kinase and protein kinase A signaling modules, but the full scope of this extensive filamentous response has not been delineated. Accordingly, we have undertaken the first systematic gene disruption and overexpression analysis of yeast filamentous growth. Standard laboratory strains of yeast are nonfilamentous; thus, we constructed a unique set of reagents in the filamentous Σ1278b strain, encompassing 3627 integrated transposon insertion alleles and 2043 overexpression constructs. Collectively, we analyzed 4528 yeast genes with these reagents and identified 487 genes conferring mutant filamentous phenotypes upon transposon insertion and/or gene overexpression. Using a fluorescent protein reporter integrated at the MUC1 locus, we further assayed each filamentous growth mutant for aberrant protein levels of the key flocculence factor Muc1p. Our results indicate a variety of genes and pathways affecting filamentous growth. In total, this filamentous growth gene set represents a wealth of yeast biology, highlighting 84 genes of uncharacterized function and an underappreciated role for the mitochondrial retrograde signaling pathway as an inhibitor of filamentous growth.     

Ingestion of Salmonella enterica Serotype Poona by a Free-Living Nematode, Caenorhabditis elegans, and Protection against Inactivation by Produce Sanitizers

Free-living nematodes are known to ingest food-borne pathogens and may serve as vectors to contaminate preharvest fruits and vegetables. Caenorhabditis elegans was selected as a model to study the effectiveness of sanitizers in killing Salmonella enterica serotype Poona ingested by free-living nematodes. Aqueous suspensions of adult worms that had fed on S. enterica serotype Poona were treated with produce sanitizers. Treatment with 20 μg of free chlorine/ml significantly (α = 0.05) reduced the population of S. enterica serotype Poona compared to results for treating worms with water (control). However, there was no significant difference in the number of S. enterica serotype Poona cells surviving treatments with 20 to 500 μg of chlorine/ml, suggesting that reductions caused by treatment with 20 μg of chlorine/ml resulted from inactivation of S. enterica serotype Poona on the surface of C. elegans but not cells protected by the worm cuticle after ingestion. Treatment with Sanova (850 or 1,200 μg/ml), an acidified sodium chlorite sanitizer, caused reductions of 5.74 and 6.34 log₁₀ CFU/worm, respectively, compared to reductions from treating worms with water. Treatment with 20 or 40 μg of Tsunami 200/ml, a peroxyacetic acid-based sanitizer, resulted in reductions of 4.83 and 5.34 log₁₀ CFU/worm, respectively, compared to numbers detected on or in worms treated with water. Among the organic acids evaluated at a concentration of 2%, acetic acid was the least effective in killing S. enterica serotype Poona and lactic acid was the most effective. Treatment with up to 500 μg of chlorine/ml, 1% hydrogen peroxide, 2,550 μg of Sanova/ml, 40 μg of Tsunami 200/ml, or 2% acetic, citric, or lactic acid had no effect on the viability or reproductive behavior of C. elegans. Treatments were also applied to cantaloupe rind and lettuce inoculated with S. enterica serotype Poona or C. elegans that had ingested S. enterica serotype Poona. Protection of ingested S. enterica serotype Poona against sanitizers applied to cantaloupe was not evident; however, ingestion afforded protection of the pathogen on lettuce. These results indicate that S. enterica serotype Poona ingested by C. elegans may be protected against treatment with chlorine and other sanitizers, although the basis for this protection remains unclear.