Learning statistical models of phenotypes using noisy labeled training data

OBJECTIVE

Traditionally, patient groups with a phenotype are selected through rule-based definitions whose creation and validation are time-consuming. Machine learning approaches to electronic phenotyping are limited by the paucity of labeled training datasets. We demonstrate the feasibility of utilizing semi-automatically labeled training sets to create phenotype models via machine learning, using a comprehensive representation of the patient medical record.

METHODS

We use a list of keywords specific to the phenotype of interest to generate noisy labeled training data. We train L1 penalized logistic regression models for a chronic and an acute disease and evaluate the performance of the models against a gold standard.

RESULTS

Our models for Type 2 diabetes mellitus and myocardial infarction achieve precision and accuracy of 0.90, 0.89, and 0.86, 0.89, respectively. Local implementations of the previously validated rule-based definitions for Type 2 diabetes mellitus and myocardial infarction achieve precision and accuracy of 0.96, 0.92 and 0.84, 0.87, respectively.We have demonstrated feasibility of learning phenotype models using imperfectly labeled data for a chronic and acute phenotype. Further research in feature engineering and in specification of the keyword list can improve the performance of the models and the scalability of the approach.

CONCLUSIONS

Our method provides an alternative to manual labeling for creating training sets for statistical models of phenotypes. Such an approach can accelerate research with large observational healthcare datasets and may also be used to create local phenotype models.

Integrated, Multi-cohort Analysis Identifies Conserved Transcriptional Signatures across Multiple Respiratory Viruses

Respiratory viral infections are a significant burden to healthcare worldwide. Many whole genome expression profiles have identified different respiratory viral infection signatures, but these have not translated to clinical practice. Here, we performed two integrated, multi-cohort analyses of publicly available transcriptional data of viral infections. First, we identified a common host signature across different respiratory viral infections that could distinguish (1) individuals with viral infections from healthy controls and from those with bacterial infections, and (2) symptomatic from asymptomatic subjects prior to symptom onset in challenge studies. Second, we identified an influenza-specific host response signature that (1) could distinguish influenza-infected samples from those with bacterial and other respiratory viral infections, (2) was a diagnostic and prognostic marker in influenza-pneumonia patients and influenza challenge studies, and (3) was predictive of response to influenza vaccine. Our results have applications in the diagnosis, prognosis, and identification of drug targets in viral infections.

•

MVS is a common transcriptional host response to respiratory viral infection

•

MVS could be used in clinics as a diagnostic and/or prognostic biomarker

•

IMS distinguishes influenza from other viral and bacterial infections

•

IMS correlates with infection symptomatology and vaccine response

Risk Stratification and Prognosis in Sepsis: What Have We Learned from Microarrays?

Sepsis mortality rates have decreased in recent years but remain unacceptably high. Risk stratification and prognostication is of particular importance because high-risk patients may benefit from earlier clinical interventions, whereas low-risk patients may benefit from not undergoing unnecessary procedures. Prognostication is currently done mostly via clinical criteria and blood lactate levels. This article summarizes the literature on the complexity of changes at the molecular level for the casual reader.

Genome-wide expression for diagnosis of pulmonary tuberculosis: a multicohort analysis

BACKGROUND

Active pulmonary tuberculosis is difficult to diagnose and treatment response is difficult to effectively monitor. A WHO consensus statement has called for new non-sputum diagnostics. The aim of this study was to use an integrated multicohort analysis of samples from publically available datasets to derive a diagnostic gene set in the peripheral blood of patients with active tuberculosis.

METHODS

We searched two public gene expression microarray repositories and retained datasets that examined clinical cohorts of active pulmonary tuberculosis infection in whole blood. We compared gene expression in patients with either latent tuberculosis or other diseases versus patients with active tuberculosis using our validated multicohort analysis framework. Three datasets were used as discovery datasets and meta-analytical methods were used to assess gene effects in these cohorts. We then validated the diagnostic capacity of the three gene set in the remaining 11 datasets.

FINDINGS

A total of 14 datasets containing 2572 samples from 10 countries from both adult and paediatric patients were included in the analysis. Of these, three datasets (N=1023) were used to discover a set of three genes (GBP5, DUSP3, and KLF2) that are highly diagnostic for active tuberculosis. We validated the diagnostic power of the three gene set to separate active tuberculosis from healthy controls (global area under the ROC curve (AUC) 0·90 [95% CI 0·85-0·95]), latent tuberculosis (0·88 [0·84-0·92]), and other diseases (0·84 [0·80-0·95]) in eight independent datasets composed of both children and adults from ten countries. Expression of the three-gene set was not confounded by HIV infection status, bacterial drug resistance, or BCG vaccination. Furthermore, in four additional cohorts, we showed that the tuberculosis score declined during treatment of patients with active tuberculosis.

INTERPRETATION

Overall, our integrated multicohort analysis yielded a three-gene set in whole blood that is robustly diagnostic for active tuberculosis, that was validated in multiple independent cohorts, and that has potential clinical application for diagnosis and monitoring treatment response. Prospective laboratory validation will be required before it can be used in a clinical setting.

FUNDING

National Institute of Allergy and Infectious Diseases, National Library of Medicine, the Stanford Child Health Research Institute, the Society for University Surgeons, and the Bill and Melinda Gates Foundation.

A comprehensive time-course-based multicohort analysis of sepsis and sterile inflammation reveals a robust diagnostic gene set

Although several dozen studies of gene expression in sepsis have been published, distinguishing sepsis from a sterile systemic inflammatory response syndrome (SIRS) is still largely up to clinical suspicion. We hypothesized that a multicohort analysis of the publicly available sepsis gene expression data sets would yield a robust set of genes for distinguishing patients with sepsis from patients with sterile inflammation. A comprehensive search for gene expression data sets in sepsis identified 27 data sets matching our inclusion criteria. Five data sets (n = 663 samples) compared patients with sterile inflammation (SIRS/trauma) to time-matched patients with infections. We applied our multicohort analysis framework that uses both effect sizes and P values in a leave-one-data set-out fashion to these data sets. We identified 11 genes that were differentially expressed (false discovery rate ≤1%, inter-data set heterogeneity P > 0.01, summary effect size >1.5-fold) across all discovery cohorts with excellent diagnostic power [mean area under the receiver operating characteristic curve (AUC), 0.87; range, 0.7 to 0.98]. We then validated these 11 genes in 15 independent cohorts comparing (i) time-matched infected versus noninfected trauma patients (4 cohorts), (ii) ICU/trauma patients with infections over the clinical time course (3 cohorts), and (iii) healthy subjects versus sepsis patients (8 cohorts). In the discovery Glue Grant cohort, SIRS plus the 11-gene set improved prediction of infection (compared to SIRS alone) with a continuous net reclassification index of 0.90. Overall, multicohort analysis of time-matched cohorts yielded 11 genes that robustly distinguish sterile inflammation from infectious inflammation.

Combined Mapping of Multiple clUsteriNg ALgorithms (COMMUNAL): A Robust Method for Selection of Cluster Number, K

In order to discover new subsets (clusters) of a data set, researchers often use algorithms that perform unsupervised clustering, namely, the algorithmic separation of a dataset into some number of distinct clusters. Deciding whether a particular separation (or number of clusters, K) is correct is a sort of ‘dark art’, with multiple techniques available for assessing the validity of unsupervised clustering algorithms. Here, we present a new technique for unsupervised clustering that uses multiple clustering algorithms, multiple validity metrics, and progressively bigger subsets of the data to produce an intuitive 3D map of cluster stability that can help determine the optimal number of clusters in a data set, a technique we call COmbined Mapping of Multiple clUsteriNg ALgorithms (COMMUNAL). COMMUNAL locally optimizes algorithms and validity measures for the data being used. We show its application to simulated data with a known K, and then apply this technique to several well-known cancer gene expression datasets, showing that COMMUNAL provides new insights into clustering behavior and stability in all tested cases. COMMUNAL is shown to be a useful tool for determining K in complex biological datasets, and is freely available as a package for R.

Prediction of neurosurgical intervention after mild traumatic brain injury using the national trauma data bank

Introduction

Patients with mild traumatic brain injury (TBI) as defined by an admission Glasgow Coma Score (GCS) of 14–15 often do not require neurosurgical interventions, but which patients will go on to require neurosurgical care has been difficult to predict. We hypothesized that injury patterns would be associated with need for eventual neurosurgical intervention in mild TBI.

Methods

The National Trauma Databank (2007–2012) was queried for patients with blunt injury and a diagnosis of TBI with an emergency department GCS of 14–15. Patients were stratified by age and injury type. Multiple logistic regression for neurosurgical intervention was run with patient demographics, physiologic variables, and injury diagnoses as dependent variables.

Results

The study included 50,496 patients, with an overall 8.8 % rate of neurosurgical intervention. Neurosurgical intervention rates varied markedly according to injury type, and were only correlated with age for patients with epidural and subdural hemorrhage. In multiple logistic regression, TBI diagnoses were predictive of need for neurosurgical interventions; moreover, after controlling for injury type and severity score, age was not significantly associated with requiring neurosurgical intervention.

Conclusions

We found that in mild TBI, injury pattern is associated with eventual need for neurosurgical intervention. Patients with cerebral contusion or subarachnoid hemorrhage are much less likely to require neurosurgical intervention, and the effects of age are not significant after controlling for other patient factors. Prospective studies should validate this finding so that treatment guidelines can be updated to better allocate ICU resources.

Using the wisdom of the crowds to find critical errors in biomedical ontologies: a study of SNOMED CT

OBJECTIVES

The verification of biomedical ontologies is an arduous process that typically involves peer review by subject-matter experts. This work evaluated the ability of crowdsourcing methods to detect errors in SNOMED CT (Systematized Nomenclature of Medicine Clinical Terms) and to address the challenges of scalable ontology verification.

METHODS

We developed a methodology to crowdsource ontology verification that uses micro-tasking combined with a Bayesian classifier. We then conducted a prospective study in which both the crowd and domain experts verified a subset of SNOMED CT comprising 200 taxonomic relationships.

RESULTS

The crowd identified errors as well as any single expert at about one-quarter of the cost. The inter-rater agreement (κ) between the crowd and the experts was 0.58; the inter-rater agreement between experts themselves was 0.59, suggesting that the crowd is nearly indistinguishable from any one expert. Furthermore, the crowd identified 39 previously undiscovered, critical errors in SNOMED CT (eg, 'septic shock is a soft-tissue infection').

DISCUSSION

The results show that the crowd can indeed identify errors in SNOMED CT that experts also find, and the results suggest that our method will likely perform well on similar ontologies. The crowd may be particularly useful in situations where an expert is unavailable, budget is limited, or an ontology is too large for manual error checking. Finally, our results suggest that the online anonymous crowd could successfully complete other domain-specific tasks.

CONCLUSIONS

We have demonstrated that the crowd can address the challenges of scalable ontology verification, completing not only intuitive, common-sense tasks, but also expert-level, knowledge-intensive tasks.

Metabolic surgery: action via hormonal milieu changes, changes in bile acids or gut microbiota? A summary of the literature

Obesity and type 2 diabetes remain epidemic problems. Different bariatric surgical techniques causes weight loss and diabetes remission to varying degrees. The underlying mechanisms of the beneficial effects of bariatric surgery are complex, and include changes in diet and behaviour, as well as changes in hormones, bile acid flow, and gut bacteria. We summarized the effects of multiple different bariatric procedures, and their resulting effects on several hormones (leptin, ghrelin, adiponectin, glucagon-like peptide 1 (GLP-1), peptide YY, and glucagon), bile acid changes in the gut and the serum, and resulting changes to the gut microbiome. As much as possible, we have tried to incorporate multiple studies to try to explain underlying mechanistic changes. What emerges from the data is a picture of clear differences between restrictive and metabolic procedures. The latter, in particular the roux-en-Y gastric bypass, induces large and distinctive changes in most measured fat and gut hormones, including early and sustained increase in GLP-1, possible through intestinal bile acid signalling. The changes in bile flow and the gut microbiome are causally inseparable so far, but new studies show that each contributes to the effects of weight loss and diabetes resolution.

The human gut microbiome: a review of the effect of obesity and surgically induced weight loss

Recent advances in parallel genomic processing and computational mapping have been applied to the native human microbial environment to provide a new understanding of the role of the microbiome in health and disease. In particular, studies of the distal gut microbiome have proposed that changes in gut microbiota are related to obesity, the metabolic syndrome, and Western diet. We examined the changes in the distal gut microbiome composition as it relates to the lean and obese phenotypes, particularly after surgical weight loss. A PubMed search of publications from January 1, 2005, through December 31, 2012, used the search terms weight, obesity, microbiome, and bariatric surgery. We included studies that provided information on subjects' weight and/or body mass index and a formal assessment of the microbiome. Certain bacteria, specifically the archaeon Methanobrevibacter smithii, have enhanced ability to metabolize dietary substrate, thereby increasing host energy intake and weight gain. With weight loss, there is a decrease in the ratio of Firmicutes to Bacteroidetes phyla. One major finding from microbial sequencing analyses after Roux-en-Y gastric bypass is the comparative overabundance of Proteobacteria in the distal gut microbiome, which is distinct from the changes seen in weight loss without Roux-en-Y gastric bypass. This review provides the practicing surgeon with (1) an update on the state of a rapidly innovating branch of clinical bioinformatics, specifically, the microbiome; (2) a new understanding of the microbiome changes after Roux-en-Y gastric bypass and weight loss; and (3) a basis for understanding further clinical applications of studies of the distal gut microbiome, such as in Crohn disease, ulcerative colitis, and infectious colitis.

A toll-like receptor 2 pathway regulates the Ppargc1a/b metabolic co-activators in mice with Staphylococcal aureus sepsis

Activation of the host antibacterial defenses by the toll-like receptors (TLR) also selectively activates energy-sensing and metabolic pathways, but the mechanisms are poorly understood. This includes the metabolic and mitochondrial biogenesis master co-activators, Ppargc1a (PGC-1α) and Ppargc1b (PGC-1β) in Staphylococcus aureus (S. aureus) sepsis. The expression of these genes in the liver is markedly attenuated inTLR2^−/− mice and markedly accentuated in TLR4^−/− mice compared with wild type (WT) mice. We sought to explain this difference by using specific TLR-pathway knockout mice to test the hypothesis that these co-activator genes are directly regulated through TLR2 signaling. By comparing their responses to S. aureus with WT mice, we found that MyD88-deficient and MAL-deficient mice expressed hepatic Ppargc1a and Ppargc1b normally, but that neither gene was activated in TRAM-deficient mice. Ppargc1a/b activation did not require NF-kβ, but did require an interferon response factor (IRF), because neither gene was activated in IRF-3/7 double-knockout mice in sepsis, but both were activated normally in Unc93b1-deficient (3d) mice. Nuclear IRF-7 levels in TLR2^−/− and TLR4^−/− mice decreased and increased respectively post-inoculation and IRF-7 DNA-binding at the Ppargc1a promoter was demonstrated by chromatin immunoprecipitation. Also, a TLR2-TLR4-TRAM native hepatic protein complex was detected by immunoprecipitation within 6 h of S. aureus inoculation that could support MyD88-independent signaling to Ppargc1a/b. Overall, these findings disclose a novel MyD88-independent pathway in S. aureus sepsis that links TLR2 and TLR4 signaling in innate immunity to Ppargc1a/b gene regulation in a critical metabolic organ, the liver, by means of TRAM, TRIF, and IRF-7.

Heme oxygenase-1 couples activation of mitochondrial biogenesis to anti-inflammatory cytokine expression

The induction of heme oxygenase-1 (HO-1; Hmox1) by inflammation, for instance in sepsis, is associated both with an anti-inflammatory response and with mitochondrial biogenesis. Here, we tested the idea that HO-1, acting through the Nfe2l2 (Nrf2) transcription factor, links anti-inflammatory cytokine expression to activation of mitochondrial biogenesis. HO-1 induction after LPS stimulated anti-inflammatory IL-10 and IL-1 receptor antagonist (IL-1Ra) expression in mouse liver, human HepG2 cells, and mouse J774.1 macrophages but blunted tumor necrosis factor-α expression. This was accompanied by nuclear Nfe2l2 accumulation and led us to identify abundant Nfe2l2 and other mitochondrial biogenesis transcription factor binding sites in the promoter regions of IL10 and IL1Ra compared with pro-inflammatory genes regulated by NF-κΒ. Mechanistically, HO-1, through its CO product, enabled these transcription factors to bind the core IL10 and IL1Ra promoters, which for IL10 included Nfe2l2, nuclear respiratory factor (NRF)-2 (Gabpa), and MEF2, and for IL1Ra, included NRF-1 and MEF2. In cells, Hmox1 or Nfe2l2 RNA silencing prevented IL-10 and IL-1Ra up-regulation, and HO-1 induction failed post-LPS in Nfe2l2-silenced cells and post-sepsis in Nfe2l2(-/-) mice. Nfe2l2(-/-) mice compared with WT mice, showed more liver damage, higher mortality, and ineffective CO rescue in sepsis. Nfe2l2(-/-) mice in sepsis also generated higher hepatic TNF-α mRNA levels, lower NRF-1 and PGC-1α mRNA levels, and no enhancement of anti-inflammatory Il10, Socs3, or bcl-x(L) gene expression. These findings disclose a highly structured transcriptional network that couples mitochondrial biogenesis to counter-inflammation with major implications for immune suppression in sepsis.

Staphylococcus aureus sepsis and mitochondrial accrual of the 8-oxoguanine DNA glycosylase DNA repair enzyme in mice

RATIONALE

Damage to mitochondrial DNA (mtDNA) by the production of reactive oxygen species during inflammatory states, such as sepsis, is repaired by poorly understood mechanisms.

OBJECTIVES

To test the hypothesis that the DNA repair enzyme, 8-oxoguanine DNA glycosylase (OGG1), contributes to mtDNA repair in sepsis.

METHODS

Using a well-characterized mouse model of Staphylococcus aureus sepsis, we analyzed molecular markers for mitochondrial biogenesis and OGG1 translocation into liver mitochondria as well as OGG1 mRNA expression at 0, 24, 48, and 72 hours after infection. The effects of OGG1 RNA silencing on mtDNA content were determined in control, tumor necrosis factor-α, and peptidoglycan-exposed rat hepatoma cells. Based on in situ analysis of the OGG1 promoter region, chromatin immunoprecipitation assays were performed for nuclear respiratory factor (NRF)-1 and NRF-2α GA-binding protein (GABP) binding to the promoter of OGG1.

MEASUREMENTS AND MAIN RESULTS

Mice infected with 10(7) cfu S. aureus intraperitoneally demonstrated hepatic oxidative mtDNA damage and significantly lower hepatic mtDNA content as well as increased mitochondrial OGG1 protein and enzyme activity compared with control mice. The infection also caused increases in hepatic OGG1 transcript levels and NRF-1 and NRF-2α transcript and protein levels. A bioinformatics analysis of the Ogg1 gene locus identified several promoter sites containing NRF-1 and NRF-2α DNA binding motifs, and chromatin immunoprecipitation assays confirmed in situ binding of both transcription factors to the Ogg1 promoter within 24 hours of infection.

CONCLUSIONS

These studies identify OGG1 as an early mitochondrial response protein during sepsis under regulation by the NRF-1 and NRF-2α transcription factors that regulate mitochondrial biogenesis.

Differential regulation of the PGC family of genes in a mouse model of Staphylococcus aureus sepsis

The PGC family of transcriptional co-activators (PGC-1alpha [Ppargc1a], PGC-1beta [Ppargc1b], and PRC [Pprc]) coordinates the upregulation of mitochondrial biogenesis, and Ppargc1a is known to be activated in response to mitochondrial damage in sepsis. Therefore, we postulated that the PGC family is regulated by the innate immune system. We investigated whether mitochondrial biogenesis and PGC gene expression are disrupted in an established model of Staphylococcus aureus sepsis both in mice with impaired innate immune function (TLR2-/- and TLR4-/-) and in wild-type controls. We found an early up-regulation of Ppargc1a and Ppargc1b post-infection (at 6 h) in WT mice, but the expression of both genes was concordantly dysregulated in TLR2-/- mice (no increase at 6 h) and in TLR4-/- mice (amplified at 6 h). However, the third family member, PRC, was regulated differently, and its expression increased significantly at 24 h in all three mouse strains (WT, TLR2-/-, and TLR4-/-). In silico analyses showed that Ppargc1a and Ppargc1b share binding sites for microRNA mmu-mir-202-3p. Thus, miRNA-mediated post-transcriptional mRNA degradation could account for the failure to increase the expression of both genes in TLR2-/- mice. The expression of mmu-mir-202-3p was measured by real-time PCR and found to be significantly increased in TLR2-/- but not in WT or TLR4-/- mice. In addition, it was found that mir-202-3p functionally decreases Ppargc1a mRNA in vitro. Thus, both innate immune signaling through the TLRs and mir-202-3p-mediated mRNA degradation are implicated in the co-regulation of Ppargc1a and Ppargc1b during inflammation. Moreover, the identification of mir-202-3p as a potential factor for Ppargc1a and Ppargc1b repression in acute inflammation may open new avenues for mitochondrial research and, potentially, therapy.

Co-regulation of nuclear respiratory factor-1 by NFkappaB and CREB links LPS-induced inflammation to mitochondrial biogenesis

The nuclear respiratory factor-1 (NRF1) gene is activated by lipopolysaccharide (LPS), which might reflect TLR4-mediated mitigation of cellular inflammatory damage via initiation of mitochondrial biogenesis. To test this hypothesis, we examined NRF1 promoter regulation by NFκB, and identified interspecies-conserved κB-responsive promoter and intronic elements in the NRF1 locus. In mice, activation of Nrf1 and its downstream target, Tfam, by Escherichia coli was contingent on NFκB, and in LPS-treated hepatocytes, NFκB served as an NRF1 enhancer element in conjunction with NFκB promoter binding. Unexpectedly, optimal NRF1 promoter activity after LPS also required binding by the energy-state-dependent transcription factor CREB. EMSA and ChIP assays confirmed p65 and CREB binding to the NRF1 promoter and p65 binding to intron 1. Functionality for both transcription factors was validated by gene-knockdown studies. LPS regulation of NRF1 led to mtDNA-encoded gene expression and expansion of mtDNA copy number. In cells expressing plasmid constructs containing the NRF-1 promoter and GFP, LPS-dependent reporter activity was abolished by cis-acting κB-element mutations, and nuclear accumulation of NFκB and CREB demonstrated dependence on mitochondrial H(2)O(2). These findings indicate that TLR4-dependent NFκB and CREB activation co-regulate the NRF1 promoter with NFκB intronic enhancement and redox-regulated nuclear translocation, leading to downstream target-gene expression, and identify NRF-1 as an early-phase component of the host antibacterial defenses.

Nitric oxide synthase-2 regulates mitochondrial Hsp60 chaperone function during bacterial peritonitis in mice

Nitric oxide synthase-2 (NOS2) plays a critical role in reactive nitrogen species generation and cysteine modifications that influence mitochondrial function and signaling during inflammation. Here, we investigated the role of NOS2 in hepatic mitochondrial biogenesis during Escherichia coli peritonitis in mice. NOS2(-/-) mice displayed smaller mitochondrial biogenesis responses than Wt mice during E. coli infection according to differences in mRNA levels for the PGC-1 alpha coactivator, nuclear respiratory factor-1, mitochondrial transcription factor-A (Tfam), and mtDNA polymerase (Pol gamma). NOS2(-/-) mice did not significantly increase mitochondrial Tfam and Pol gamma protein levels during infection in conjunction with impaired mitochondrial DNA (mtDNA) transcription, loss of mtDNA copy number, and lower State 3 respiration rates. NOS2 blockade in mitochondrial-GFP reporter mice disrupted Hsp60 localization to mitochondria after E. coli exposure. Mechanistically, biotin-switch and immunoprecipitation studies demonstrated NOS2 binding to and S-nitros(yl)ation of Hsp60 and Hsp70. Specifically, NOS2 promoted Tfam accumulation in mitochondria by regulation of Hsp60-Tfam binding via S-nitros(yl)ation. In hepatocytes, site-directed mutagenesis identified (237)Cys as a critical residue for Hsp60 S-nitros(yl)ation. Thus, the role of NOS2 in inflammation-induced mitochondrial biogenesis involves both optimal gene expression for nuclear-encoded mtDNA-binding proteins and functional regulation of the Hsp60 chaperone that enables their importation for mtDNA transcription and replication.

A new in vivo microvascular preparation of the hamster ovary

Ovarian function in the cycling female is intimately related to and dependent upon significant microvascular regulation and restructuring. To enable investigation of the microvascular determinants of ovarian function, we present an in vivo preparation of the golden hamster ovary. The preparation does not compromise the ovarian vascular supply. The viability and responsiveness of the preparation were confirmed by quantifying arteriolar responses to vasoactive agents in 17 hamsters. Small surface arterioles (mean diameter 15-16 microns) responded with statistically significant changes in diameter to adenosine and oxygen and showed significant, dose-dependent constriction in response to norepinephrine and the NO synthase inhibitor L-NAME. Other key findings included extremely high microvascular permeability that varied with the day of the estrous cycle and functionally significant architectural features of the utero-ovarian vascular network. Potential applications of the preparation include elucidation of the role of the microvasculature in follicular development and luteal regression, investigation of utero-ovarian crossregulation, and development of a model for the study of ovarian angiogenesis and vascular regression.

Alteration of testicular microvascular pressures during venous pressure elevation

We have addressed the hypothesis that varicocele-related infertility is caused in part by a pressure-induced disturbance of testicular convective transport that upsets the testicular hormonal environment and thus impairs spermatogenesis. The left testis of the hamster [pentobarbital sodium (Nembutal), 70 mg/kg ip] was prepared for microcirculatory observations. Testicular venous pressure was acutely elevated by ligating collateral routes of venous outflow and partially occluding, via a snare, the main venous outflow distal to the pampiniform plexus. Simultaneous direct pressure measurements (servo-null method) were made to monitor venous pressure elevation and quantify resulting pressure and diameter changes in the arterial feed to the testis and in postcapillary venules. The data show that over 90% of the venous pressure elevation (VPE) was transmitted to the postcapillary venules. VPE affected intravascular pressures throughout the testis microvasculature; on average, capsular artery pressure increased by 83% of the VPE, although part of this increase was due to a rise in systemic arterial pressure. Vasoconstriction helped to buffer the pressure rise in the capsular artery, probably at the expense of flow amplitude. Yet the vasoconstriction was ineffective in preventing a rise in exchange vessel pressure. These data suggest that microvascular fluid exchange may be dramatically altered in varicocele, upsetting the hormonal and paracrine environment of the testis, and hence, impairing physiological regulation of gametogenesis.

Limitations of methods of osmometry: measuring the osmolality of biological fluids

Osmometry is an important tool in the investigation of biological phenomena, and commercially available instruments for freezing point and vapor pressure osmometry can determine the osmolality of solutions quickly and inexpensively. However, accurate measurements of osmolality using these techniques require that the solutions have specific characteristics, and that measurements do not exceed the limitations inherent to each method or instrument. The thermodynamic principles underlying osmometry constrain the range and accuracy of each measurement method, and these must be considered in establishing the usefulness of each technique. This paper addresses the principles and limitations of routine osmometry techniques. We begin by discussing definitions of osmolality and the thermodynamic concepts of solute-solvent systems that are central to understanding osmometry of biological (i.e., aqueous) solutions. We then explore the application of various methods of measuring osmolality, the nature of errors introduced by overextension or misapplication of osmometry techniques, and the interpretation of data in the literature acquired by various methods and protocols.

Microvascular pressure distribution in the hamster testis

Convective transport is a critical element in the regulation of steroidogenesis and spermatogenesis in the testis. Steroid hormones are distributed to their target cells within seminiferous tubules via interstitial fluid. The movement of interstitial fluid and lymph, which transports protein hormones and many of the substrates required for spermatogenesis and steroidogenesis, is driven by capillary filtration. Despite the importance of convective transport in testicular function, however, the mechanisms regulating transvascular exchange in the testis are unknown. As a first step in understanding this process, we measured directly the microvascular hydrostatic pressure distribution in the hamster testis (pentobarbital sodium, 70 mg/kg ip). Using a servo-null transducer, intravascular pressure was measured in all vessel types accessible beneath the surface of the testis of 19 animals. Systemic arterial pressure averaged 89 +/- 2 (SE) mmHg. The most significant observations were that mean capillary pressure was extremely low (10.1 +/- 0.8 mmHg) and remarkably constant (range 8.2-13.3 mmHg), despite a 45 mmHg range in systemic mean arterial pressure among the animals observed. The maintenance of a low hydrostatic pressure in testicular capillaries may serve to sustain fluid filtration at a rate that prevents washout of essential solutes while preserving convective transport. Unfortunately, the anatomical and functional characteristics that determine this unique microvascular environment may also expose the testis to significant pathological risks. For example, the large pre- to postcapillary resistance ratio observed suggests that testicular capillaries must be highly susceptible to increases in venous pressure.(ABSTRACT TRUNCATED AT 250 WORDS)

Spatial heterogeneity in striated muscle arteriolar tone, cell flow, and capillarity

Spatial variations in microvascular function are described at two tissue sites in hamster cremaster muscle (pentobarbital sodium, 70 mg/kg ip). Arterioles observed include terminal arterioles and their feeding vessels, termed capillary network controllers (CNC). Although terminal arterioles at both sites had similar maximum diameters and cell flows, those at site I were significantly more constricted at rest (2.7 +/- 0.3 vs. 5.1 +/- 0.3 microns at site II) and showed lower resting flows (19.0 +/- 5.5 vs. 174 +/- 34 cells/s at site II). There were no spatial differences in CNC maximal parameters or CNC resting tone, yet CNC resting flow at site II (798 +/- 118 cells/s) significantly exceeded the value at site I (460 +/- 85 cells/s). At rest, median capillary cell flow at site I (3.3 cells/s) was half that at site II (6.3 cells/s). During hyperemia, perfused capillary segment length per unit volume was 84% greater at site I and estimated tissue erythrocyte content nearly double that at site II. Thus significant spatial differences in microvascular function exist in cell flow and vessel tone among terminal arterioles, in cell flow among CNC, and in capillarity and indices of capillary exchange.

The effect of dexamethasone on chronic pulmonary oxygen toxicity in infant mice

The effect of dexamethasone (0.1, 1, and 5 mg/kg/d given subcutaneously from d 14-18) was tested in infant mice continuously exposed from birth to either humidified air or 80% oxygen. Dexamethasone significantly decreased lung wet wt (p less than 0.01), lung water (p less than 0.021), lung dry wt, protein, and DNA (p less than 0.001) in both air- and oxygen-exposed animals. Dexamethasone, however, had no effect on lung compliance measured after animals were killed on d 18. It also had no effect on the increase in the blood-air barrier thickness or decrease in the blood-air exchange surface area seen in the 80% oxygen-exposed mice. Dexamethasone decreased thymus gland wt (p less than 0.001), body wt gain (p less than 0.001), brain wt (p less than 0.001), and lung lymphocytes (p less than 0.05) in both air- and oxygen-exposed animals. The effect of 1 mg/kg and 5 mg/kg of the drug could not be differentiated. During the 4 d of drug administration, one air- and one oxygen-exposed animal died; both received 5 mg/kg/d of dexamethasone; microscopic and culture evidence of infection was not found. If dexamethasone causes similar effects in human infants with bronchopulmonary dysplasia, it should be used with great caution even for short-term clinical management.

Arteriolar control of capillary cell flow in striated muscle

This study tests the hypothesis that capillary perfusion is controlled in groups rather than at the level of the individual capillary. We measured cell flux (using cells labeled with substituted tetramethyl rhodamine isothiocyanate, XRITC) and vessel diameter in adjoining arterioles of the terminal vasculature of hamster cremaster muscle (Nembutal, 70 mg/kg i.p.) during rest and hyperemia (10(-4) M adenosine). In terminal arterioles (TAs), 32 of 68 vessels showed cell flux increases from rest to hyperemia exceeding 25 times (i.e., 47% of TAs were relatively unperfused at rest). In vessels feeding TAs (TAFs), 33 of 95 (34%) were relatively unperfused at rest. Cell flux heterogeneity in TAFs decreased significantly by 27% from rest to hyperemia; the corresponding decrease (16%) in TAs was not significant. Thus, unperfused TAFs are present in a proportion which reflects capillary recruitment in hamster cremaster (Sarelius et al, Am J Physiol 1981;241:H317) while TAs are not, and TAFs independently modulate flow distribution distally while TAs do not. The data therefore support the conclusion that TAFs control cell flow in the distal microvasculature. Analysis of normalized ranked maximal diameters showed that TAFs unperfused at rest tend to be the smaller vessels at any tissue site.

Use of flow cytometry to extend and improve in vivo determinations of microvessel hematocrit and cell flux

The use of fluorescent-erythrocyte tracer methods for estimating microvascular variables has been extended and improved by using flow cytometry. With cell tracer methods, the precision of estimates of cell flux, hematocrit, etc. is dependent on the labeled-cell fraction (p). In larger arterioles, high cell flux and the requirement to identify each labeled cell mandate the use of very low labeled-cell fractions. Precise microvascular estimates require high total cell counts in the estimation of p. This is not practical to do manually, but can be accomplished by flow cytometry. Determinations of p in experimental samples were made by both flow cytometry and microscopy. These values were compared to each other and to calculated values. In the range 0.0045 less than p less than 0.075, where manual counting statistics give CV(p) less than 0.12, counts by microscopy and flow cytometry were not significantly different. This confirms that flow cytometry identifies the same cell population as seen in vivo. Flow cytometry counts will resolve p greater than 0.00001 for rhodamine-labeled-cell fractions and p greater than 0.0002 for fluorescein-labeled-cell fractions. Rhodamine-labeled-cell fraction samples analyzed when fresh or after 1-3 weeks storage in RPMI 1640 media (GIBCO) gave p values which agreed to within 4%. We identify the use of rhodamine-labeled cells as a simple yet powerful method of rare-event analysis for many in vivo applications.