Hybridization and gene expression: Beyond differentially expressed genes

Gene expression has a key role in reproductive isolation, and studies of hybrid gene expression have identified mechanisms causing hybrid sterility. Here, we review the evidence for altered gene expression following hybridization and outline the mechanisms shown to contribute to altered gene expression in hybrids. Transgressive gene expression, transcending that of both parental species, is pervasive in early generation sterile hybrids, but also frequently observed in viable, fertile hybrids. We highlight studies showing that hybridization can result in transgressive gene expression, also in established hybrid lineages or species. Such extreme patterns of gene expression in stabilized hybrid taxa suggest that altered hybrid gene expression may result in hybridization-derived evolutionary novelty. We also conclude that while patterns of misexpression in hybrids are well documented, the understanding of the mechanisms causing misexpression is lagging. We argue that jointly assessing differences in cell composition and cell-specific changes in gene expression in hybrids, in addition to assessing changes in chromatin and methylation, will significantly advance our understanding of the basis of altered gene expression. Moreover, uncovering to what extent evolution of gene expression results in altered expression for individual genes, or entire networks of genes, will advance our understanding of how selection moulds gene expression. Finally, we argue that jointly studying the dual roles of altered hybrid gene expression, serving both as a mechanism for reproductive isolation and as a substrate for hybrid ecological adaptation, will lead to significant advances in our understanding of the evolution of gene expression.

Different complex regulatory phenotypes underlie hybrid male sterility in divergent rodent crosses

Hybrid incompatibilities are a critical component of species barriers and may arise due to negative interactions between divergent regulatory elements in parental species. We used a comparative approach to identify common themes in the regulatory phenotypes associated with hybrid male sterility in two divergent rodent crosses, dwarf hamsters and house mice. We investigated three potential characteristic regulatory phenotypes in hybrids including the propensity towards over or underexpression relative to parental species, the influence of developmental stage on the extent of misexpression, and the role of the sex chromosomes on misexpression phenotypes. In contrast to near pervasive overexpression in hybrid house mice, we found that misexpression in hybrid dwarf hamsters was dependent on developmental stage. In both house mouse and dwarf hamster hybrids, however, misexpression increased with the progression of spermatogenesis, although to varying extents and with potentially different consequences. In both systems, we detected sex-chromosome specific overexpression in stages of spermatogenesis where inactivated X chromosome expression was expected, but the hybrid overexpression phenotypes were fundamentally different. Importantly, misexpression phenotypes support the presence of multiple histological blocks to spermatogenesis in dwarf hamster hybrids, including a potential role of meiotic stalling early in spermatogenesis. Collectively, we demonstrate that while there are some similarities in hybrid regulatory phenotypes of house mice and dwarf hamsters, there are also clear differences that point towards unique mechanisms underlying hybrid male sterility in each system. Our results highlight the potential of comparative approaches in helping to understand the importance of disrupted gene regulation in speciation.

Evolution of parent-of-origin effects on placental gene expression in house mice

The mammalian placenta is a hotspot for the evolution of genomic imprinting, a form of gene regulation that involves the parent-specific epigenetic silencing of one allele. Imprinted genes are central to placental development and are thought to contribute to the evolution of reproductive barriers between species. However, it is unclear how rapidly imprinting evolves or how functional specialization among placental tissues influences the evolution of imprinted expression. We compared parent-of-origin expression bias across functionally distinct placental layers sampled from reciprocal crosses within three closely related lineages of mice ( Mus ). Using genome-wide gene expression and DNA methylation data from fetal and maternal tissues, we developed an analytical strategy to minimize pervasive bias introduced by maternal contamination of placenta samples. We corroborated imprinted expression at 42 known imprinted genes and identified five candidate imprinted genes showing parent-of-origin specific expression and DNA methylation. Paternally-biased expression was enriched in the labyrinth zone, a layer specialized in nutrient transfer, and maternally-biased genes were enriched in the junctional zone, which specializes in modulation of maternal physiology. Differentially methylated regions were predominantly determined through epigenetic modification of the maternal genome and were associated with both maternally- and paternally-biased gene expression. Lastly, comparisons between lineages revealed a small set of co-regulated genes showing rapid divergence in expression levels and imprinted status in the M. m. domesticus lineage. Together, our results reveal important links between core functional elements of placental biology and the evolution of imprinted gene expression among closely related rodent species.

A natural variation-based screen in mouse cells reveals USF2 as a regulator of the DNA damage response and cellular senescence

Cellular senescence is a program of cell cycle arrest, apoptosis resistance, and cytokine release induced by stress exposure in metazoan cells. Landmark studies in laboratory mice have characterized a number of master senescence regulators, including p16INK4a, p21, NF-κB, p53, and C/EBPβ. To discover other molecular players in senescence, we developed a screening approach to harness the evolutionary divergence between mouse species. We found that primary cells from the Mediterranean mouse Mus spretus, when treated with DNA damage to induce senescence, produced less cytokine and had less-active lysosomes than cells from laboratory Mus musculus. We used allele-specific expression profiling to catalog senescence-dependent cis-regulatory variation between the species at thousands of genes. We then tested for correlation between these expression changes and interspecies sequence variants in the binding sites of transcription factors. Among the emergent candidate senescence regulators, we chose a little-studied cell cycle factor, upstream stimulatory factor 2 (USF2), for molecular validation. In acute irradiation experiments, cells lacking USF2 had compromised DNA damage repair and response. Longer-term senescent cultures without USF2 mounted an exaggerated senescence regulatory program-shutting down cell cycle and DNA repair pathways, and turning up cytokine expression, more avidly than wild-type. We interpret these findings under a model of pro-repair, anti-senescence regulatory function by USF2. Our study affords new insights into the mechanisms by which cells commit to senescence, and serves as a validated proof of concept for natural variation-based regulator screens.

Distinct genetic origins of eumelanin intensity and barring patterns in cichlid fishes

Pigment patterns are incredibly diverse across vertebrates and are shaped by multiple selective pressures from predator avoidance to mate choice. A common pattern across fishes, but for which we know little about the underlying mechanisms, is repeated melanic vertical bars. In order to understand genetic factors that modify the level or pattern of vertical barring, we generated a genetic cross of 322 F2 hybrids between two cichlid species with distinct barring patterns, Aulonocara koningsi and Metriaclima mbenjii. We identify 48 significant quantitative trait loci that underlie a series of seven phenotypes related to the relative pigmentation intensity, and four traits related to patterning of the vertical bars. We find that genomic regions that generate variation in the level of eumelanin produced are largely independent of those that control the spacing of vertical bars. Candidate genes within these intervals include novel genes and those newly-associated with vertical bars, which could affect melanophore survival, fate decisions, pigment biosynthesis, and pigment distribution. Together, this work provides insights into the regulation of pigment diversity, with direct implications for an animal's fitness and the speciation process.

Genetic basis of ecologically relevant body shape variation among four genera of cichlid fishes

Divergence in body shape is one of the most widespread and repeated patterns of morphological variation in fishes and is associated with habitat specification and swimming mechanics. Such ecological diversification is the first stage of the explosive adaptive radiation of cichlid fishes in the East African Rift Lakes. We use two hybrid crosses of cichlids (Metriaclima sp. × Aulonocara sp. and Labidochromis sp. × Labeotropheus sp., >975 animals total) to determine the genetic basis of body shape diversification that is similar to benthic-pelagic divergence across fishes. Using a series of both linear and geometric shape measurements, we identified 34 quantitative trait loci (QTL) that underlie various aspects of body shape variation. These QTL are spread throughout the genome, each explaining 3.2-8.6% of phenotypic variation, and are largely modular. Further, QTL are distinct both between these two crosses of Lake Malawi cichlids and compared to previously identified QTL for body shape in fishes such as sticklebacks. We find that body shape is controlled by many genes of small effect. In all, we find that convergent body shape phenotypes commonly observed across fish clades are most likely due to distinct genetic and molecular mechanisms.

Adaptive structural and functional evolution of the placenta protects fetal growth in high-elevation deer mice

Environmental hypoxia challenges female reproductive physiology in placental mammals, increasing rates of gestational complications. Adaptation to high elevation has limited many of these effects in humans and other mammals, offering potential insight into the developmental processes that lead to and protect against hypoxia-related gestational complications. However, our understanding of these adaptations has been hampered by a lack of experimental work linking the functional, regulatory, and genetic underpinnings of gestational development in locally adapted populations. Here, we dissect high-elevation adaptation in the reproductive physiology of deer mice (Peromyscus maniculatus), a rodent species with an exceptionally broad elevational distribution that has emerged as a model for hypoxia adaptation. Using experimental acclimations, we show that lowland mice experience pronounced fetal growth restriction when challenged with gestational hypoxia, while highland mice maintain normal growth by expanding the compartment of the placenta that facilitates nutrient and gas exchange between gestational parent and fetus. We then use compartment-specific transcriptome analyses to show that adaptive structural remodeling of the placenta is coincident with widespread changes in gene expression within this same compartment. Genes associated with fetal growth in deer mice significantly overlap with genes involved in human placental development, pointing to conserved or convergent pathways underlying these processes. Finally, we overlay our results with genetic data from natural populations to identify candidate genes and genomic features that contribute to these placental adaptations. Collectively, these experiments advance our understanding of adaptation to hypoxic environments by revealing physiological and genetic mechanisms that shape fetal growth trajectories under maternal hypoxia.

The evolution of widespread recombination suppression on the dwarf hamster (Phodopus) X chromosome

The X chromosome of therian mammals shows strong conservation among distantly related species, limiting insights into the distinct selective processes that have shaped sex chromosome evolution. We constructed a chromosome-scale de novo genome assembly for the Siberian dwarf hamster (Phodopus sungorus), a species reported to show extensive recombination suppression across an entire arm of the X chromosome. Combining a physical genome assembly based on shotgun and long-range proximity ligation sequencing with a dense genetic map, we detected widespread suppression of female recombination across ∼65% of the Phodopus X chromosome. This region of suppressed recombination likely corresponds to the Xp arm, which has previously been shown to be highly heterochromatic. Using additional sequencing data from two closely related species (P. campbelli and P. roborovskii), we show that recombination suppression on Xp appears to be independent of major structural rearrangements. The suppressed Xp arm was enriched for several transposable element families and de-enriched for genes primarily expressed in placenta, but otherwise showed similar gene densities, expression patterns, and rates of molecular evolution when compared to the recombinant Xq arm. Phodopus Xp gene content and order was also broadly conserved relative to the more distantly related rat X chromosome. These data suggest that widespread suppression of recombination has likely evolved through the transient induction of facultative heterochromatin on the Phodopus Xp arm without major changes in chromosome structure or genetic content. Thus, substantial changes in the recombination landscape have so far had relatively subtle influences on patterns of X-linked molecular evolution in these species.

X chromosome-dependent disruption of placental regulatory networks in hybrid dwarf hamsters

Embryonic development in mammals is highly sensitive to changes in gene expression within the placenta. The placenta is also highly enriched for genes showing parent-of-origin or imprinted expression, which is predicted to evolve rapidly in response to parental conflict. However, little is known about the evolution of placental gene expression, or if divergence of placental gene expression plays an important role in mammalian speciation. We used crosses between two species of dwarf hamsters (Phodopus sungorus and Phodopus campbelli) to examine the genetic and regulatory underpinnings of severe placental overgrowth in their hybrids. Using quantitative genetic mapping and mitochondrial substitution lines, we show that overgrowth of hybrid placentas was primarily caused by genetic differences on the maternally inherited P. sungorus X chromosome. Mitochondrial interactions did not contribute to abnormal hybrid placental development, and there was only weak correspondence between placental disruption and embryonic growth. Genome-wide analyses of placental transcriptomes from the parental species and first- and second-generation hybrids revealed a central group of co-expressed X-linked and autosomal genes that were highly enriched for maternally biased expression. Expression of this gene network was strongly correlated with placental size and showed widespread misexpression dependent on epistatic interactions with X-linked hybrid incompatibilities. Collectively, our results indicate that the X chromosome is likely to play a prominent role in the evolution of placental gene expression and the accumulation of hybrid developmental barriers between mammalian species.

Chromosome-scale assemblies reveal the structural evolution of African cichlid genomes

BACKGROUND

African cichlid fishes are well known for their rapid radiations and are a model system for studying evolutionary processes. Here we compare multiple, high-quality, chromosome-scale genome assemblies to elucidate the genetic mechanisms underlying cichlid diversification and study how genome structure evolves in rapidly radiating lineages.

RESULTS

We re-anchored our recent assembly of the Nile tilapia (Oreochromis niloticus) genome using a new high-density genetic map. We also developed a new de novo genome assembly of the Lake Malawi cichlid, Metriaclima zebra, using high-coverage Pacific Biosciences sequencing, and anchored contigs to linkage groups (LGs) using 4 different genetic maps. These new anchored assemblies allow the first chromosome-scale comparisons of African cichlid genomes. Large intra-chromosomal structural differences (∼2-28 megabase pairs) among species are common, while inter-chromosomal differences are rare (<10 megabase pairs total). Placement of the centromeres within the chromosome-scale assemblies identifies large structural differences that explain many of the karyotype differences among species. Structural differences are also associated with unique patterns of recombination on sex chromosomes. Structural differences on LG9, LG11, and LG20 are associated with reduced recombination, indicative of inversions between the rock- and sand-dwelling clades of Lake Malawi cichlids. M. zebra has a larger number of recent transposable element insertions compared with O. niloticus, suggesting that several transposable element families have a higher rate of insertion in the haplochromine cichlid lineage.

CONCLUSION

This study identifies novel structural variation among East African cichlid genomes and provides a new set of genomic resources to support research on the mechanisms driving cichlid adaptation and speciation.

An allelic series at pax7a is associated with colour polymorphism diversity in Lake Malawi cichlid fish

Despite long-standing interest in the evolution and maintenance of discrete phenotypic polymorphisms, the molecular genetic basis of such polymorphism in the wild is largely unknown. Female sex-associated blotched colour polymorphisms found in cichlids of Lake Malawi, East Africa, represent a highly successful polymorphic phenotype, found and maintained in four genera across the geographic expanse of the lake. Previously, we identified an association with an allelic variant of the paired-box transcription factor gene pax7a and blotched colour morphs in Lake Malawi cichlid fishes. Although a diverse range of blotched phenotypes are present in Lake Malawi cichlid species, they all appeared to result from an allele of pax7a that produces increased levels of transcript. Here, we examine the developmental and genetic basis of variation among blotched morphs. First, we confirm that pax7a-associated blotch morphs result primarily from modulation of melanophore development and survival. From laboratory crosses and natural population studies, we identify at least three alleles of pax7a associated with discrete subtypes of blotched morphs, in addition to the ancestral pax7a allele. Genotypes at pax7a support initial evolution of a novel pax7a allele to produce the blotched class of morphs, followed by subsequent evolution of that pax7a blotched allele to produce additional alleles associated with discrete colour morphs. Variant alleles of pax7a produce different levels of pax7a transcript, correlating with pigmentation phenotype at the cellular level. This naturally selected allelic series should serve as a case study for understanding the molecular genetic control of pax7a expression and the evolution of sex-associated alleles.

A simple tool for mortality prediction in burns patients: APACHE III score and FTSA

Prediction of outcome for patients with major thermal injury is important to inform clinical decision making, alleviate individual suffering and improve hospital resource allocation. Age and burn size are widely accepted as the two largest contributors of mortality amongst burns patients. The APACHE (Acute Physiology and Chronic Health Evaluation) III-j score, which incorporates patient age, is also useful for mortality prediction, of intensive care populations. Validation for the burns specific cohort is unclear. A retrospective cohort study was performed on patients admitted to the Intensive Care Unit (ICU) via the Victorian Adult Burns Service (VABS), to compare observed mortality with burns specific markers of illness severity and APACHE III-j score. Our primary aim was to develop a mortality prediction tool for the burns population. Between January 1, 2002 and December 31, 2008, 228 patients were admitted to the ICU at The Alfred with acute burns. The mean age was 45.6 years and 81% (n=184) were male. Patients had severe injuries: the average percent TBSA (total body surface area) was 28% (IQR 10-40) and percent FTSA (full thickness surface area) was 18% (IQR 10-25). 86% (n=197) had airway involvement. Overall mortality in the 7-year period was 12% (n=27). Non-survivors were older, had larger and deeper burns, a higher incidence of deliberate self-harm, higher APACHE III-j scores and spent less time in hospital (but similar time in ICU), compared with survivors. Independent risk factors for death were percent FTSA (OR 1.03, 95% CI 1.01-1.05, p=0.01) and APACHE III-j score (OR 1.04, 95% CI 1.02-1.07, p<0.001). Mortality prediction based on both of these variables in combination was more specific than either individual variable alone (AUROC 0.85, 95% CI 0.79-0.92). Likelihood of death for patients with severe thermal injury can be predicted with accuracy from APACHE III-j score and percent FTSA. Prospective validation of our model on different burn populations is necessary.

Jagged1 (JAG1) mutations in Alagille syndrome: increasing the mutation detection rate

Alagille syndrome (AGS) is caused by heterozygous mutations in JAG1, and mutations have been previously reported in about 70% of patients who meet clinical diagnostic criteria. We studied a cohort of 247 clinically well-defined patients, and using an aggressive and sequential screening approach we identified JAG1 mutations in 94% of individuals. Mutations were found in 232 out of 247 patients studied and 83 of the mutations were novel. This increase in the mutation rate was accomplished by combining rigorous clinical phenotyping, with a combination of mutation detection techniques, including fluorescence in situ hybridization (FISH), genomic and cDNA sequencing, and quantitative PCR. This higher rate of mutation identification has implications for clinical practice, facilitating genetic counseling, prenatal diagnosis, and evaluation of living-related liver transplant donors. Our results suggest that more aggressive screening may similarly increase the rate of mutation detection in other dominant and recessive disorders.

Enhancement of human immunodeficiency virus type 1-specific CD4 and CD8 T cell responses in chronically infected persons after temporary treatment interruption

Immunologic and virologic outcomes of treatment interruption were compared for 5 chronically human immunodeficiency virus (HIV)-infected persons who have maintained antiretroviral therapy-mediated virus suppression, as compared with 5 untreated controls. After a median interruption of 55 days of therapy accompanied by rebound of virus, reinitiated therapy in 4 of 5 subjects resulted in suppression of 98.86% of plasma virus load by 21-33 days and no significant decrease in CD4 T cell percentage from baseline. Increased T helper responses against HIV-1 p24 antigen (P=. 014) and interferon-gamma-secreting CD8 T cell responses against HIV-1 Env (P=.004) were present during interruption of therapy and after reinitiation of treatment. The remaining subject whose treatment was interrupted did not resume treatment and continued to have a low virus load (<1080 HIV-1 RNA copies/mL) and persistent antiviral cell-mediated responses. In summary, cellular immunity against autologous HIV-1 has the potential to be acutely augmented in association with temporary treatment interruption in chronically infected persons.

Lack of long-term effects of in utero exposure to zidovudine among uninfected children born to HIV-infected women. Pediatric AIDS Clinical Trials Group Protocol 219/076 Teams

CONTEXT

With the success of zidovudine chemoprophylaxis for prevention of perinatal transmission of the human immunodeficiency virus (HIV), an increasing number of HIV-exposed but uninfected children will have in utero exposure to zidovudine and other antiretroviral drugs.

OBJECTIVE

To evaluate the long-term effects of in utero exposure to zidovudine vs placebo among a randomized cohort of uninfected children.

DESIGN

Prospective cohort study based on data collected during Pediatric AIDS Clinical Trials Group Protocol 076, a perinatal zidovudine HIV prevention trial, and Protocol 219, a long-term observational protocol.

SETTING

Pediatric research clinics in the United States.

PATIENTS

Two hundred thirty-four uninfected children born to 230 HIV-infected women enrolled in Protocol 076 and followed up through February 28, 1997, in Protocol 219 (122 in the zidovudine group and 112 in the placebo group).

MAIN OUTCOME MEASURES

Physical growth measurements, immunologic parameters, cognitive/developmental function, occurrence of neoplasms, and mortality data assessed every 6 months for children younger than 24 months and yearly thereafter or as clinically indicated. Baseline echocardiogram and funduscopic evaluations were collected before 36 months of age.

RESULTS

Median age of children at time of last follow-up visit was 4.2 years (range, 3.2-5.6 years). There were no significant differences between children exposed to zidovudine and those who received placebo in terms of sequential data on lymphocyte subsets; weight, height, and head circumference z scores; and cognitive/developmental function. No deaths or malignancies occurred. Two children (both exposed to zidovudine) are being followed up for abnormal, unexplained ophthalmic findings. One child exposed to zidovudine had a mild cardiomyopathy on echocardiogram at the age of 48 months; the child is clinically asymptomatic.

CONCLUSIONS

No adverse effects were observed in HIV-uninfected children with in utero and neonatal exposure to zidovudine followed up for as long as 5.6 years. Continued prospective evaluations of children born to HIV-infected women who are exposed to antiretroviral or immunotherapeutic agents are critical to assess the long-term safety of interventions that prevent perinatal HIV transmission.

Influence of pre-adolescent diet on quantitative ultrasound measurements of the calcaneus in young adult women

Nongenetic determinants of quantitative ultrasound (QUS) properties of the bone remain to be identified. The purpose of this study was to determine relationships between early adolescent diet and QUS bone measurements taken in young adulthood. Subjects were participants in the 10-year longitudinal National Heart, Lung, and Blood Institute Growth and Health Study (NGHS). QUS parameters measured at the calcaneus in a convenience subsample of 63 18- to 19-year-old black and white women were correlated with dietary data collected when the subjects were aged 9-11 years. We hypothesized that pre-adolescent intake of calcium, magnesium, vitamin C and protein, nutrients known to be associated with bone development, would be associated with QUS measurements in young women. Stepwise multiple regression analysis, controlling for race, height and weight, demonstrated that pre-adolescent intake of calcium and magnesium were positively related to QUS parameters (calcium with broadband ultrasound attenuation, and magnesium with speed of sound and bone velocity). Our findings suggest that pre-adolescent diet may be associated with bone properties as measured by ultrasound. Further investigations of this relationship may yield a deeper understanding of the impact of diet on skeletal development. The small size of the convenience sample used for the analysis precludes stronger inferences at this time.

Aberrant neutrophil trafficking and metabolic oscillations in severe pyoderma gangrenosum

Having previously associated metabolic oscillations with cell locomotion, we hypothesized that patients with abnormalities in neutrophil trafficking may display aberrant intracellular oscillations. A pyoderma gangrenosum patient exhibiting aberrant leukocyte trafficking in vivo and skin ulceration without infection was identified. This patient's neutrophils constitutively overexpressed and clustered the leukocyte integrins CR3 and CR4 and failed to display appropriate integrin-to-GPI receptor interactions. Increased levels of tyrosine phosphorylation were observed. NAD(P)H oscillations, which are sinusoidal in normals, were chaotic with multiple frequency components in this patient's neutrophils. Normal cell shape and sinusoidal NAD(P)H oscillations were restored by providing a pulsed electric field to drive metabolic oscillations and by temperature reduction. N-acetyl-D-glucosamine disrupted CR3 clusters and sinusoidal NAD(P)H oscillations returned. Anecdotal reports suggest that local hypothermia is clinically useful for this patient. These data define the first metabolic oscillation-associated disease and suggest that pyoderma gangrenosum can be classified as a dynamical disease at the cellular level.

Aberrant integrin (CR4; alpha(x)beta2; CD11c/CD18) oscillations on neutrophils in a mild form of pyoderma gangrenosum

We have previously shown that the beta2 integrins CR3 and CR4 physically and functionally interact with urokinase receptors (uPAR) on neutrophil plasma membranes in an oscillatory fashion. In this study we have analyzed neutrophils from patient SC, a 34 y old African American female, with aberrant skin window results and recurrent perianal abscesses and pretibial lesions diagnosed as pyoderma gangrenosum. Although untreated migrating normal neutrophils exhibited 20 s sinusoidal oscillations in CR4-uPAR proximity, neutrophils from SC demonstrated a faster oscillation (10 s) in the form of a flyback sawtooth wave. This waveform mimicked that observed for normal neutrophils treated with subsaturating doses of the kinase inhibitors staurosporine, genistein, and erbstatin. As beta2 integrins are regulated by phosphorylation, we tested the hypothesis that the aberrant CR4-uPAR proximity oscillations seen in SC's neutrophils are due to defective kinase activity that might be balanced by a decrease in phosphatase activity. When SC's cells are exposed to subsaturating concentrations of the phosphatase inhibitor pervanadate, this caused the CR4-uPAR oscillations to become sinusoidal in shape with a 20 s period, as seen in normal migrating neutrophils. Although SC's neutrophils were deficient in spontaneous and N-formyl-methionyl-leucyl-phenylalanine-induced polarization, 0.5 microM pervanadate returned cell polarization to nearly normal levels, thus paralleling the acquisition of normal receptor interactions. Inasmuch as SC's cellular phenotype is mimicked by kinase inhibitors and corrected by phosphatase inhibitors, we suggest that a mutation(s) affecting the kinetics of intracellular signaling enzymes, but not blocking the pathway per se, may be responsible for this clinical state.

Thermal inactivation of bovine immunodeficiency virus

Cell-associated bovine immunodeficiency virus (BIV) and cell-free BIV were subjected to increasing temperatures, including pasteurization conditions. To determine the effect of heat treatment on BIV viability, reverse transcriptase activity and infectivity of the heat-treated virus were assessed. BIV was inactivated by heating to 47 degrees C for 30 min and by low- and high-temperature pasteurization conditions.

The effect of finasteride, a 5 alpha-reductase inhibitor, on scalp skin testosterone and dihydrotestosterone concentrations in patients with male pattern baldness

The effects of the 5 alpha-reductase inhibitor, finasteride, on scalp skin testosterone (T) and dihydrotestosterone (DHT) levels were studied in patients with male pattern baldness. In a double blind study, male patients undergoing hair transplantation were treated with oral finasteride (5 mg/day) or placebo for 28 days. Scalp skin biopsies were obtained before and after treatment for measurement of T and DHT by high pressure liquid chromatography-RIA. In 10 male subjects studied at baseline, mean (+/- SEM) DHT levels were significantly higher in bald (7.37 +/- 1.24 pmol/g) compared to hair-containing (4.20 +/- 0.65 pmol/g) scalp, whereas there was no difference in mean T levels at baseline. In bald scalp from 8 patients treated with finasteride, the mean DHT concentration decreased from 6.40 +/- 1.07 pmol/g at baseline to 3.62 +/- 0.38 pmol/g on day 28. Scalp T levels increased in 6 of 8 subjects treated with finasteride. Finasteride decreased the mean serum DHT concentration from 1.36 +/- 0.18 nmol/L (n = 8) at baseline to 0.46 +/- 0.10 nmol/L on day 28 and had no effect on serum T. There were no significant changes in scalp or serum T or DHT in placebo-treated patients. In this study, male subjects treated with 5 mg/day finasteride for 4 weeks had significantly decreased concentrations of DHT in bald scalp, resulting in a mean level similar to the baseline levels found in hair-containing scalp.

Staphylococcal infections in childhood dermatomyositis--association with the development of calcinosis, raised IgE concentrations and granulocyte chemotactic defect

There is a high incidence of staphylococcal infection in children with dermatomyositis, which is limited to those children who either already have or subsequently develop calcinosis. Of 15 children followed up for 3-10 years after diagnosis, all nine who developed calcinosis had infections with Staphylococcus aureus compared with none of six without calcinosis. Of these nine, the occurrence of staphylococcal infections before calcinosis was observed in four, suggested by history in two, and unclear in three children. Granulocyte chemotaxis to Staphylococcus aureus was more severely depressed in those children with calcinosis, whereas those without calcinosis did not differ significantly from controls. The chemotactic defect was due to a serum factor (patients' serum depressed control chemotaxis and control serum corrected the patients' chemotaxis). The nine children with calcinosis also had significantly higher serum IgE concentrations than non-atopic age matched controls; the six without calcinosis did not differ from controls. The increased IgE concentrations appeared to develop after staphylococcal infection and before calcinosis. Two of five patients with calcinosis had increased antistaphylococcal IgE antibodies; neither of the two patients without calcinosis had such increased antibodies. This suggests preceding immunological differences in patients with dermatomyositis who do and do not subsequently develop calcinosis, either increasing susceptibility to Staphylococcus aureus infection or potentially resulting from such infections.

Focal scleroderma and severe cardiomyopathy. Patient report and brief review

A 21-month-old infant presented with simultaneous localized scleroderma and severe cardiomyopathy with heart failure. Cardiac abnormalities and serological changes (positive rheumatoid factor assay, elevated IgM and IgG levels, and elevated erythrocyte sedimentation rate) reverted to normal with prednisone therapy, and there was substantial, though incomplete, resolution of her skin changes during the same period. To our knowledge, this is the first patient with definite, clinically significant cardiac involvement associated with focal scleroderma. The possibility of internal organ involvement, including cardiac involvement, must be considered with focal scleroderma as well as with progressive systemic sclerosis.

Inhibition of two enzymes in de novo purine nucleotide synthesis by triciribine phosphate (TCN-P)

We previously reported that triciribine (tricyclic nucleoside, TCN, NSC-154020), after phosphorylation in cultured CCRF-CEM human leukemic lymphoblasts inhibited de novo purine nucleotide synthesis, GTP more than ATP [Moore et al. Biochem. Pharmac. 38, 4037 (1989)]. To determine the enzymes inhibited, triciribine phosphate (TCN-P, NSC-280594) was tested in dialyzed extracts of the cells. A new assay for glycinamide ribotide (GAR) synthesis was based on incorporation of [14C]glycine into GAR as a ribose-containing compound retained on boronyl gel columns. Glutamine, phosphoribosyl pyrophosphate (PRPP), ATP and glycine were required for the two-step sequence of glutamine:amidophosphoribosyltransferase (EC 2.4.2.14) and phosphoribosylamine-glycine ligase (EC 6.3.4.13). When PRPP was near the normal intracellular concentration (0.1 mM), 1.2 mM TCN-P inhibited GAR synthesis by 71-95%. To permit separate assay of the ligase step, 6-diazo-5-oxo-L-norleucine was used to inhibit amidophosphoribosyltransferase and phosphoribosylamine (PRA) was supplied in situ by chemical reaction of ribose-5-phosphate and ammonia (as ammonium acetate). The ligase was not inhibited by TCN-P. Thus, TCN-P inhibits amidophosphoribosyltransferase; it acts as an analog of the purine nucleotides which regulate this first committed step of de novo purine biosynthesis by an allosteric feedback mechanism. The measured intracellular concentration (0.1 mM) of PRPP was not changed in cells treated with TCN. IMP dehydrogenase (EC 1.1.1.205), the first de novo step committed to guanosine nucleotide synthesis, was also tested. It was inhibited by TCN-P, competitively with IMP, 66% at 1.2 mM TCN-P and 8 microM IMP. The degree of inhibition of these two enzymes was sufficient to account for the effects on purine nucleotide biosynthesis observed in intact cells treated with TCN.

Inhibition of CCRF-CEM human leukemic lymphoblasts by triciribine (tricyclic nucleoside, TCN, NSC-154020). Accumulation of drug in cells and comparison of effects on viability, protein synthesis and purine synthesis

The experimental antineoplastic agent triciribine (tricyclic nucleoside, TCN) is known to be activated to its phosphate TCN-P by adenosine kinase and to inhibit cell growth, purine nucleotide synthesis, and incorporation of amino acids into proteins. Our objective in this paper was to compare these effects in intact cells of a human cell line as a prerequisite to describing in a companion paper [Moore et al., Biochem. Pharmac. 38, 4045 (1989)] more detailed enzymic studies of their interrelationships. TCN treatment inhibited cloning of CCRF-CEM human leukemic lymphoblasts 50% at concentrations of 6, 30, and 90 microM with 8-day, 8-hr, and 2-hr exposures respectively. However, 6-20% of the cells survived exposure to 200 microM TCN for 24 hr. The intracellular formation of TCN-P from TCN was rapid, concentrative and essentially complete, but TCN-P did not exceed about 1.4 mM (1.4 nmol/10(6) cells) at 200 microM TCN. In cells exposed to 50 microM TCN for 1.25 to 24 hr, formate incorporation into ATP and GTP was inhibited the most rapidly and strongly; pools of ATP and GTP were decreased as much as 40% (as compared with controls); and incorporation of formate into RNA purines was inhibited as much as 65%. Incorporation of leucine into protein was more moderately inhibited up to 40%, apparently in proportion to the concentration of intracellular TCN-P, rather than of the TCN in the medium. These inhibitions occurred most rapidly during the first 2-4 hr and increased only gradually thereafter, whereas cloning ability was inhibited more slowly and uniformly over a longer time period. No one of these metabolic effects by itself showed a clear correlation with the loss of viability. The incorporation of formate into formylglycinamide ribotide (FGAR, when accumulated at a blockage by azaserine) was inhibited drastically by TCN. The rate of incorporation of hypoxanthine into ATP was increased by TCN, whereas incorporation into GTP was decreased. Thus, the principal sites of inhibition of purine synthesis by TCN-P were shown in these intact cells to be at a step prior to synthesis of FGAR in the de novo pathway and also at an additional site between IMP and GTP.

Biochemical pharmacology of N-acetyl-N-(methylcarbamoyloxy)-N'-methylurea (caracemide; NSC-253272)

Preclinical pharmacologic studies of caracemide [N-acetyl-N-(methylcarbamoyloxy)-N'-methylurea; CAR] have demonstrated a marked instability of this compound in the presence of either phosphate buffer (pH 7.4) or human plasma. Using [1-14C-acetyl]CAR and [3H-methylcarbamoyloxy]CAR, three CAR degradation products were identified: product A, N-(methylcarbamoyloxy)acetamide; product B: N-(methylcarbamoyloxy)-N'-methylurea; and product C: N-hydroxy-N'-methylurea. CAR degradation in human plasma was demonstrated by high-performance liquid chromatography (HPLC) to occur in a time- and temperature-dependent manner. A 30-min incubation (37 degrees) of CAR (10(-4) M) with human plasma resulted in degradation of more than 55% of parent compound; at 1 hr, more than 75% of original CAR was degraded. Incubation of [1-14C-acetyl]CAR with rat brain homogenate resulted in the formation of 14CO2. This reaction was partially inhibited by coincubation with physostigmine (10(-3) M). CAR inhibited acetylcholinesterase activity in neuroblastoma cells with an IC50 of 14 microM. In mechanism of action studies, CAR was found to inhibit ribonucleotide reductase activity but only at nine times the IC50 of hydroxyurea. In contrast to hydroxyurea, CAR was found to be non-cell-cycle phase-specific and non-cross-resistant with two CHO cell lines resistant to hydroxyurea. These data demonstrate the instability of CAR; moreover, they suggest that its mechanism of cytotoxicity is distinctly different from that of hydroxyurea and that the neurotoxicity associated with CAR administration may be caused in part by inhibition of acetylcholinesterase activity.

Epidemic Shiga bacillus dysentery in Central Africa

An outbreak of dysentery began late in 1979 in Central Africa and spread to involve a major portion of Zaire as well as Rwanda and Burundi. We traveled to a mission hospital in northeast Zaire during the epidemic and isolated Shigella dysenteriae, type 1, from most of the patients studied. All isolates were resistant to ampicillin, tetracycline, chloramphenicol, sulfathiazole, and streptomycin but sensitive to trimethoprim-sulfamethoxazole. Antimicrobial resistance was transferable to Escherichia coli, and at least three plasmids were identified in the donor Shigella isolates by using agarose gel electrophoresis. One was coded for ampicillin, tetracycline, and chloramphenicol resistance while a second conferred resistance to ampicillin and chloramphenicol but not tetracycline. A third large plasmid of approximately 120 megadaltons could not be transferred to E. coli recipients. All S. dysenteriae isolates yielded identical kinetic growth curves when analyzed on the Abbot MS-2 Research System. This is the most extensive outbreak of dysentery caused by S. dysenteriae reported since the Central American epidemic of 1969, and the first epidemic caused by a strain resistant to ampicillin.

Inhibition of ribonucleotide reductase by caracemide

Caracemide, a new antitumor agent now in clinical trial, was tested against partially purified ribonucleotide reductase from rat Novikoff tumor. It was found to be an active inhibitor, about one-ninth as effective as hydroxyurea.

Studies of N-hydroxy-N'-aminoguanidine derivatives by nitrogen-15 nuclear magnetic resonance spectroscopy and as ribonucleotide reductase inhibitors

Hydroxyguanidine, with the imino group of guanidine and the hydroxyamino group of hydroxyurea, has functional groups believed to be important for both anticancer and antiviral activities (Adamson, R.H. Nature (London) 1972, 236, 400-401). Three new N-hydroxy-N'-aminoguanidine derivatives have been synthesized and found to be 20-30 times more active than the hydroxyguanidine itself as inhibitors of ribonucleotide reductase from rat Novikoff tumors (Tai, W.A.; Lai, M.M.; Lien, E.J. "Novel N-Hydroxyguanidine Derivatives as Antiviral Agents", North American Medicinal Chemistry Symposium, University of Toronto, Toronto, Canada, June 20-24, 1982; Abstr, p 144). The character of the tautomeric equilibria, the pKa values, and the protonation sites of these hydroxyguanidine derivatives have been determined by 15N NMR spectroscopy.

Pharmacological and biochemical interactions of N-(phosphonacetyl)-L-aspartate and 5-fluorouracil in beagles

N-(Phosphonacetyl)-L-aspartate (PALA) and 5-fluorouracil (FUra) are both antimetabolites that affect the biosynthetic pathways of pyrimidines. To determine whether these two drugs exhibit synergistic pharmacological or biochemical interactions, we determined the pharmacological and biochemical parameters of PALA and [14C]FUra in 14 beagle dogs which received i.v. bolus administrations of either the single agents or the drug combination. The pharmacokinetic parameters of PALA (four dogs, 20 mg/kg) in plasma, cerebrospinal fluid, and urine were not changed by FUra (10 mg/kg, 30 min after PALA). The pharmacokinetics of [2-(14)C]FUra (six dogs, 10 mg/kg, 20 muCi/kg) was characterized by higher FUra plasma concentrations after pretreatment with PALA (20 mg/kg, 30 min before FUra); this led to a significantly larger area under the drug concentration-time curve, a decreased volume of distribution, and a reduced clearance rate and was associated with higher cerebrospinal fluid concentrations of FUra. The FUra plasma and cerebrospinal fluid half-lives, however, were not significantly altered by PALA. The biochemical determinants of PALA and FUra activity were studied in intestinal mucosa, liver, thymus, spleen, and bone marrow of four dogs. Although the activity of the target enzyme of PALA, L-aspartate carbamoyltransferase, in tissue extracts was decreased at least 50% at 18 to 24 hr after PALA administration (50 mg/kg), the uridine nucleotide pools remained remarkably stable. Intracellular FUra concentrations were not influenced by PALA. The incorporation of 5-fluorouridine triphosphate into RNA was enhanced in intestinal mucosa and liver. In other tissues, however, fluorouridine nucleotide concentrations were not affected by PALA. Free 5-fluorodeoxyuridine monophosphate had the highest concentration in liver and was detectable in all tissues, but it was not altered by PALA treatment. Our results show that the pharmacological and biochemical events after FUra exposure are marginally modulated by PALA in normal dogs. If sensitive tumors with a higher degree of interaction between the two drugs could be identified, limited toxicity to normal tissues can be expected.

Aspartate carbamoyltransferase activity, drug concentrations, and pyrimidine nucleotides in tissue from patients treated with N-(phosphonacetyl)-L-aspartate

Biopsy specimens were obtained from patients treated with N-(phosphonacetyl)-L-aspartate (PALA) in a phase I clinical trial. Activities of aspartate carbamoyltransferase (ACTase), the target enzyme, in ten specimens before treatment varied from 0.4 to 1.7 units/mg. PALA was measured in protein-free extracts of thirteen specimens by inhibition of rat ACTase. At 1.5 to 145 hr after doses of 1 to 6 g/m2, PALA concentrations were 0.9 to 89 micrograms/g; at 4 hr or later the tissue concentrations were similar to those in plasma (five samples). The observed inhibition of ACTase (17-87%) correlated with the PALA concentrations. Pyrimidine nucleotides were decreased (relative to purine nucleotides) in nine to ten specimens, by 16-72%. ACTase partially purified from human spleen had a Km for carbamoyl phosphate of 20.6 microM and the Ki for PALA was 0.011 microM. The results suggest that inhibition of ACTase by PALA affects the concentration of pyrimidine nucleotides in human tumors in a dose-dependent manner.

N-(Phosphonacetyl)-L-aspartate inhibition of the enzyme complex of pyrimidine biosynthesis

The inhibition of aspartate carbamoyltransferase (ACTase) from rat Novikoff tumor by N-(phosphonacetyl)-L-aspartate (PALA) was studied in a substrate mixture permitting endogenous synthesis of carbamoyl phosphate. Among the components required for carbamoyl phosphate synthetase activity, ATP, Mg(C2H3O2)2 and KCl interfered with inhibition by PALA (with added carbamoyl phosphate). The inhibition was also decreased when the concentration of partially purified enzyme was increased. In the system dependent on carbamoyl phosphate synthetase, the 50% inhibitory concentration of PALA was lower than that in the same mixture plus 0.2 mM carbamoyl phosphate, but higher than in the usual simple assay mixture with 0.2 mM carbamoyl phosphate.

Penetration of N-(phosphonacetyl)-L-aspartate into human central nervous system and intracerebral tumor

Cerebrospinal fluid (CSF) was obtained from five patients by lumbar puncture and from two patients by Ommaya reservoir tap after the i.v. administration of the antitumor agent N-(phosphonacetyl)-L-aspartate (PALA). PALA was quantified enzymatically by inhibition of the target enzyme, aspartate carbamoyltransferase. After a 1-hr infusion of PALA, its CSF concentration steadily rose until the eighth hr, at which time it was 12 to 40% of concurrent plasma concentration. PALA concentration then declined more gradually in CSF than in plasma, and CSF concentrations exceeded plasma concentrations by 24 hr. PALA concentration X time product in CSF was 12 to 25% of that in plasma. PALA was infused i.v. for 30 to 60 min into eight patients undergoing surgical resection if intracerebral tumors. Its concentration in intracerebral tumor was greater than or comparable to concentration in temporalis muscle in four of six patients from whom muscle was obtained. The PALA concentration in edematous brain tissue was consistently lower than the concentration in tumor or muscle. In a patient undergoing occipital lobectomy, the PALA concentration in brain was inversely proportional to the distance from the tumor. PALA reached concentrations in intracerebral tumor that appeared to be similar to concentrations reported previously in s.c. tumors, although biopsy techniques and conditions differed.

Pharmacological disposition of N-(phosphonacetyl)-L-aspartate in humans

The pharmacological disposition of N-(phosphonacetyl)-L-aspartate (PALA), an antitumor transition state analog currently in clinical trial, has been studied in 19 patients after i.v. administration of the agent at doses ranging from 800 to 5000 mg/sq m; PALA in biological specimens was assayed enzymatically, advantage being taken of its inhibition of L-aspartate carbamoyltransferase (EC 2.1.3.2) PALA disappeared from plasma biexponentially, with an average terminal t 1/2 of 5.3 hr. It was excreted in the urine unchanged; the 24-hr cumulative excretion was 85% of the administered dose. The average total clearance was 1.60 ml/kg/min and was linearly related to creatinine clearance, suggesting that in humans PALA is essentially cleared by glomerular filtration. The apparent volume of distribution of PALA was 309 ml/kg, approximately the sulfate space in humans. PALA penetrated into the central nervous system only to a limited extent. Tumor L-aspartate carbamoyltransferase activity was also measured before and 1.5 hr to 6 days after PALA administration; in all eight studies, a notable decrease in enzyme activity was observed.

Regional deficiency of secretory IgA in a patient with combined immunodeficiency of the ADA deficient type

The IgA system in a patient with SCID and ADA deficiency showed heterogeneity. Serum IgA and stool secretory IgA (SIgA) levels were normal, but with altered kappa/lambda and A1/A2 subclass ratios; IgA in saliva and urine was deficient. Amounts of secretory component were normal. Jejunal and rectal biopsies showed prominent lymphonodular hyperplasia, but no cells containing IgA. A normal serum IgA level therefore does not always predict an intact secretory IgA system.

Enzymatic assay for the antitumor agent-N-(phosphonacetyl)-L-aspartic acid (PALA)

A method is described for the assay of N-(phosphonacetyl)-L-aspartic acid in plasma or urine, based on inhibition of partially purified aspartate carbamoyltransferase from rat liver. Concentrations of N-(phosphonacetyl)-L-aspartic acid as low as 0.1 microgram/ml can be detected. Plasma disappearance and urinary excretion curves from one patient are shown as examples of the application of the method.

Comparison of thioredoxin reductases from Novikoff ascites hepatoma cells and normal liver of rats

Adult rat liver contained variant forms of thioredoxin reductase with isoelectric points at pH 4.9 and at approximately pH 4.7 compared to pH 5.1 for the enzyme from Novikoff ascites hepatoma. Fetal and regenerating liver contained only the form with the isoelectric point at pH 4.9. All three enzymes precipitated with and were inhibited by a rabbit antibody to purified enzyme from Novikoff tumor.

Effect of the solvent-dependent conformational system of hydroxyureas on predicted vs. observed log P

Calculated and observed log P values are reported and compared with in vivo and in vitro biological action (L1210 leukemia ILS % and ribonucleotide reductase ID50) for hydroxyurea, the 1-N methyl and ethyl, and the 3-N ethyl, n-propyl, isopropyl, n-butyl, tert-butyl, phenyl, and p-chlorophenyl analogues. The log P values were calculated via the method of Hansch and Leo from literature f values and the observed log P values were obtained by direct determination after equilibration between octanol and water. Calculations of log P for hydroxyurea were found to be appreciably more hydrophilic than the values obtained experimentally. Differences in calculated and observed log P (delta log P) for the substituted analogues were lowest with the 1-N and the bulky 3-N substituents and greatest with the 3-N-substituted straight-chain analogues (delta log P = 0.70). Different structural species were observed by infrared spectroscopy in dry octanol vs. octanol after water equilibration and drying, and this is proposed as due to changes in conformational equilibrium in the hydroxyurea systems. Differences between calculated and observed log P are explained via the stabilization of internally hydrogen-bonded conformers in the case of 1-N or bulky 3-N analogues or destabilization of various conformers allowing maximal interactions with solvent or water which is the case with straight chain 3-N analogues.

Solvent-dependent conformational system of hydroxyureas in octanol-water and in inhibition of ribonucleotide reductase

The carbonyl nitrogen IR absorption region of selected hydroxyureas was examined in octanol. The spectra indicated appreciably different absorptions prior to and after equilibration with water. Stability studies on the hydroxyureas indicated that the absorption differences were not due to chemical decomposition in the equilibration process but were due to solvent conformation and/or tautomeric dependency of the system. Preliminary in vitro inhibition of the enzyme ribonucleoside diphosphate reductase by selected hydroxyureas also indicated an apparent solvent dependency, which may involve the conformational and/or tautomeric properties of these agents. The implications of the solvent-dependent conformation-tautomeric system observed are discussed in relation to this property in the biological action of these agents.

4-methyl-5-amino-1-formylisoquinoline thiosemicarbazone, a second-generation antineoplastic agent of the alpha-(N)-heterocyclic carboxaldehyde thiosemicarbazone series

4-Methyl-5-amino-1-formylisoquinoline thiosemicarbazone (MAIQ-1) was studied to determine its potential for clinical trail as a second-generation antineoplastic agent of the alpha-(N)-heterocyclic carboxaldehyde thiosemicarbazone class. MAIQ-1 was shown to be among the most potent known inhibitors of the major target for the expression of antineoplastic activity by this class of agents, the enzyme ribonucleoside diphosphate reductase, requiring only 0.06 micronM for 50% inhibition. This potency at the enzymatic level was consistent with its antineoplastic activity against the murine neoplasms Sarcoma 180, Leukemia L1210, Leukemia P388, and the B16 melanoma. The acetylation of the 5-amino group of the model substrate 5-amino-1,4-dimethylisoquinoline was lower than that of 5-amino-1-methylisoquinoline when incubated with acetyl-coenzyme A and rat liver homogenate. This finding suggests that the presence of the 4-methyl function offers steric hinderance to enzymatic substitution of the adjacent 5-amino group. In vivo metabolism of MAIQ-1 in mice, studied with [3'-14C]MAIQ-1 showed that relatively slow excretion of this agent occurred, since the cumulative urinary excretion of radioactivity was only 35% in 48 HR. About 51% of excreted urinary radioactivity was present in chromatograms in an area corresponding to the iron chelate of MAIQ-1, and only a minor quantity of material migrating like acetylated MAIQ-1 was present in urine, a finding consistent with enzymatic data with liver homogenates. The results indicate that MAIQ-1 has the antineoplastic activity, enzyme inhibitory potency, and relative resistance to metabolic inactivation required of an agent of this class for clinical trials.

Purification of thioredoxin reductase from the Novikoff rat tumor

Thioredoxin reductase (E.C.1.6.4.5.) has been purified to about 95% homogeneity from the Novikoff ascites rat tumor. The enzyme contained two subunits of approximately 58,000 daltons, with one FAD per subunit. The amino acid analysis is reported. An immunoadsorbent was prepared and used for affinity chromatography in order to improve the yield of the enzyme.