Adherence Patterns and Adherence-Related DNA Sequences in Escherichia coli Isolates from Children with and without Diarrhea in S?o Paulo City,Brazil

The correlation between various adherence patterns and adherence-related DNA sequences in Escherichia coli isolates from 1- to 4-year-old children with and without diarrhea in São Paulo, Brazil, was evaluated. A total of 1,801 isolates obtained from 200 patients and 200 age-matched controls were studied. The adherence patterns found were classified as diffuse, aggregative, aggregative in a 6-h assay, aggregative predominantly in coverslips, localized, localized-like, and noncharacteristic. In general, the DNA sequences used as probes showed excellent specificities (>93%), but their sensitivities varied. Thus, the results of bioassays and assays with DNA probes normally used to search for adherent E. coli did not correlate well, and the best method for the identification of these organisms in the clinical research setting remains controversial. Isolates presenting diffuse adherence or hybridizing with the related daaC probe, or both, were by far the most frequent in patients (31.5, 26.0, and 23.0%, respectively), followed by isolates presenting aggregative adherence or hybridizing with the related EAEC probe, or both (21.5, 13.0, and 10.5%, respectively). None of the different combinations of adherence patterns and adherence-related DNA sequences found were associated with acute diarrhea.The first step in the establishment of the diarrheal diseases caused by the various categories of diarrheagenic Escherichia coli is adherence to epithelial cells of the intestinal mucosa. In vitro assays with eukaryotic cell lines (HeLa and HEp-2 cells) have identified three distinct adherence patterns among fecal isolates of E. coli: localized, diffuse, and aggregative (37, 38, 41). Localized adherence (LA) is characterized by formation of bacterial microcolonies on a restricted area(s) of the cell surface, while diffuse adherence (DA) is the scattered attachment of bacteria over the whole surface of the cell (41). The pattern of aggregative adherence (AA) consists of bacterial attachment to the cells and the intervening cell growth surface in a stacked brick-like lattice (37).The LA pattern was first detected in strains classified as enteropathogenic E. coli (EPEC) among serogroups associated with outbreaks of infantile diarrhea (41). Although E. coli strains exhibiting DA (DAEC) have been isolated at similar frequencies from feces of infants and young children with acute diarrhea and nondiarrheic controls in some populations (3, 10, 11, 14, 18), they were significantly associated with diarrhea in other settings (1, 17, 24, 29, 33). E. coli strains showing AA, termed enteroaggregative E. coli (EAEC), have been linked to sporadic persistent diarrhea (3, 4, 7, 10, 13, 26, 27, 44) and to outbreaks of diarrhea in both developing and developed countries (8, 12, 28, 43). However, the role of EAEC in acute diarrhea is still controversial: some studies have shown a correlation (7, 23, 25, 27, 34, 37), but others (1, 3, 6, 10, 11, 13–15, 17, 18, 24, 26, 29, 33, 44) have not.DNA probes derived from adherence-related sequences have been constructed (2, 5, 16, 31, 36) and used in hybridization assays for the detection of the different established and putative categories of diarrheagenic E. coli in many epidemiological studies.We evaluated the relationship between the LA, DA, and AA patterns and hybridization with adherence-related DNA sequences and tested children 1 to 4 years old with and without acute diarrhea for the presence of adherent E. coli strains.     

The influence of genomic context on mutation patterns in the human genome inferred from rare variants

Understanding patterns of spontaneous mutations is of fundamental interest in studies of human genome evolution and genetic disease. Here, we used extremely rare variants in humans to model the molecular spectrum of single-nucleotide mutations. Compared to common variants in humans and human–chimpanzee fixed differences (substitutions), rare variants, on average, arose more recently in the human lineage and are less affected by the potentially confounding effects of natural selection, population demographic history, and biased gene conversion. We analyzed variants obtained from a population-based sequencing study of 202 genes in >14,000 individuals. We observed considerable variability in the per-gene mutation rate, which was correlated with local GC content, but not recombination rate. Using >20,000 variants with a derived allele frequency ≤10⁻⁴, we examined the effect of local GC content and recombination rate on individual variant subtypes and performed comparisons with common variants and substitutions. The influence of local GC content on rare variants differed from that on common variants or substitutions, and the differences varied by variant subtype. Furthermore, recombination rate and recombination hotspots have little effect on rare variants of any subtype, yet both have a relatively strong impact on multiple variant subtypes in common variants and substitutions. This observation is consistent with the effect of biased gene conversion or selection-dependent processes. Our results highlight the distinct biases inherent in the initial mutation patterns and subsequent evolutionary processes that affect segregating variants.Mutation is one of the most fundamental processes in biology. It is the ultimate source of genetic variation and one of the driving forces of evolution. Mutation also plays a significant role in the etiology of human diseases. There is considerable interest in understanding the underlying pattern and molecular spectrum of spontaneous mutations. Historically, two approaches were applied to estimate the single-nucleotide mutation rate in humans. The first analyzes divergent sites between humans and another species, typically chimpanzee. According to Kimura''s neutral theory, the majority of substitutions are neutral and therefore the extent of between-species divergence can be used to estimate the neutral mutation rate (Kimura 1983). Many groups have applied this approach to estimate the spontaneous mutation rate in humans (Drake et al. 1998; Nachman and Crowell 2000; Kumar and Subramanian 2002; Silva and Kondrashov 2002). However, several forces, including natural selection, biased gene conversion (BGC), and demographic history, can alter fixation probabilities and reshape the spectrum and genomic distribution of between-species substitution patterns. A second, more direct approach, pioneered by Haldane (1935), uses incidence rates of dominant disorders in humans to estimate the mutation rate (Sommer 1995; Sommer and Ketterling 1996; Kondrashov 2003; Lynch 2010). This approach, however, is limited by the fact that only a small subset of new mutations manifest as disease variants (Nachman 2004).The mutation rates from these studies represent a genome-wide average. However, there is extensive variability among different genes or genomic regions in both between-species divergence and within-species diversity (Wolfe et al. 1989; Nachman and Crowell 2000; Sachidanandam et al. 2001; Smith and Lercher 2002; Kondrashov 2003; Hodgkinson et al. 2009). This suggests that spontaneous mutation rates are not constant throughout the genome, although the reasons behind this variability are unclear.Local nucleotide composition is a frequently studied feature that could contribute to mutation rate variability. One study showed that AT > GC (an A base replaced with a G or a T base replaced with a C) common variants segregate at a higher frequency in regions with higher GC content (Webster et al. 2003), and others similarly reported increased fixation bias toward GC base pairs in GC-rich regions (Lercher and Hurst 2002a; Lercher et al. 2002). However, analyses of GC content and variant patterns often reported contradicting findings. For example, while some studies showed that GC content is positively correlated with both divergence rates between humans and chimpanzee (Smith et al. 2002; Webster et al. 2003; Arndt and Hwa 2005; Duret and Arndt 2008) and within-human nucleotide diversity (Sachidanandam et al. 2001; Hellmann et al. 2005), another study found a negative correlation (Cai et al. 2009). Furthermore, while some studies reported increasing GC > AT substitution rates with increasing GC content (Smith et al. 2002; Webster et al. 2003), others showed a decrease (Arndt and Hwa 2004; Duret and Arndt 2008). These inconsistencies could be partly explained by differences in the allele frequency, and therefore the evolutionary time scale of the variants analyzed in different studies. Consequently the observed patterns could be the result of confounding factors, such as selection and demography, instead of alterations in the actual mutation rate.Recombination is known to influence patterns of common variation and substitution rates. Correlations between recombination rate and nucleotide diversity or between species substitution rates have been observed in humans (Nachman et al. 1998; Nachman 2001; Lercher and Hurst 2002b; Hellmann et al. 2003, 2005; Spencer et al. 2006; Duret and Arndt 2008; Cai et al. 2009; Lohmueller et al. 2011), Drosophila (Begun and Aquadro 1992; Begun et al. 2007; Kulathinal et al. 2008), and several plant species (Dvorak et al. 1998; Kraft et al. 1998; Stephan and Langley 1998; Tenaillon et al. 2004). Three major theories exist to explain these observations. First, recombination may be directly mutagenic, leading to increased mutation rates in regions of high recombination and thus higher diversity (Lercher and Hurst 2002b; Hellmann et al. 2003, 2008). Second, while background selection and selective sweeps reduce haplotype diversity, recombination generates new haplotypes by shuffling variants onto different backgrounds, thereby maintaining diversity in regions of high recombination rates (Kaplan et al. 1989; Charlesworth et al. 1993, 1995; Hudson and Kaplan 1995; Nachman 2001). A third explanation is BGC, a recombination-associated process that preferentially repairs AT/GC mismatches produced during recombination to GC bases, leading to preferential fixation of GC alleles (for review, see Duret and Galtier 2009). Over time, the observed effect of BGC can mimic that of natural selection, leading to an excess of “weak” (W) A/T bases converted to “strong” (S) G/C bases as if the latter were under positive selection (Berglund et al. 2009; Galtier et al. 2009; Necsulea et al. 2011). The reports hypothesizing a mutagenic effect of recombination relied on common variants and substitutions (Lercher and Hurst 2002b; Hellmann et al. 2003, 2008). Several lines of evidence argue against the mutagenic recombination theory and instead suggest that a selection-dependent mechanism or BGC can explain the observed correlation between diversity and recombination rate (Duret and Arndt 2008; Berglund et al. 2009; Galtier et al. 2009; Lohmueller et al. 2011).Previous studies using common variants within humans and substitutions between humans and chimpanzees are effectively dealing with mutations accumulated over many generations. Their patterns, therefore, reflect the cumulative influence of many processes, including natural selection, population demographic history, and BGC. A major challenge in the field is to elucidate the extent to which these forces alter the distribution of variants over time and to distinguish their relative contributions. To minimize the effects of selection, many studies restrict their analysis to noncoding regions of the genome. However, widespread signatures of recent positive selection, even within supposedly neutral regions (Williamson et al. 2007), suggest that noncoding regions may also be influenced by selection.Rare variants represent a newly available and expanding resource that can overcome some of these limitations. Rare variants are relatively young, predominantly because they are the result of recent mutation events. Therefore, rare variants are typically less affected by population demographic history or natural selection (Messer 2009). Furthermore, as BGC acts only on variants after they have arisen in the population (Duret and Galtier 2009), it does not influence innate mutation rates. Rare variants, therefore, are an appropriate resource for studying the spectrum and genomic distribution of mutations while minimizing the potentially confounding influences. In addition, while family-based whole-genome sequencing has begun to identify de novo mutations that provide more direct measures of mutation rates (The 1000 Genomes Project Consortium 2010; Conrad et al. 2011; Campbell et al. 2012; Kong et al. 2012), the identified mutations sparsely cover the genome. For example, if whole-genome sequencing of each parent-offspring trio yields ∼40 de novo mutations (Conrad et al. 2011), 500 such trios would need to be sequenced to accumulate roughly 20,000 mutations. These mutations, however, would occur once per 150 kb on average, and the data would lack the spatial resolution necessary to detect the effect of local genomic context on a finer scale.We studied a set of rare variants discovered via targeted resequencing of 202 genes in >14,000 unrelated individuals. We analyzed the per-gene mutation rate as well as the probability of each site to contain a variant of a specific subtype relative to local GC content, recombination rate, and recombination hotspots. In order to compare mutation rate inferences based on rare variants with those obtained by within- and between-species data, we compared rare variant patterns to common variant data from The 1000 Genomes Project Consortium and substitution sites between humans and chimpanzee. These three variant classes cover different evolutionary time scales, and the differences between them allow us to examine the distinct influence of genomic context on the initial mutation process, the subsequent rise of some mutations to become common variants, and eventual fixation.