Precise statistical distribution theory then determines the reliable P-values for making the decision. design-island runs in two phases, namely first phase and refinement phase. In the first phase, it identifies islands at different locations of the chromosome and to determine the stretches of those islands, and carries out statistical analysis using a probing window.

This leads to the identification of some ‘putative GIs’ having varying sizes and locations in the chromosome that are identifiable with P-values generated using Monte–Carlo tests carried out at variable locations of the probing window with a fixed size. Following the first phase, the refinement phase commences, which takes random samples of genomic segments excluding the regions detected in the first phase. Some of the putative GIs identified in the first phase are further

refined into smaller segments containing NVP-BKM120 supplier horizontally acquired genes in the refinement phase. design-island was implemented on the chromosomes of three completely sequenced genomes of V. cholerae under study in order to identify the putative GIs in their genomes. In the first phase, design-island was run using P0=0.05, word size of 4 and initial window size of 5000 with consequent window increment of 500. Two hundred randomly selected fragments CYC202 nmr were tested for each window with a sliding window 500. In the refinement phase or the second phase design-island was run with the same parameter values as used in the first phase, except for the initial window size, which was reduced to 2000 and the sliding window increased to 1000. The statistical analysis in the refinement phase is similar to that used in the first phase except the P0 was set to 0.001. The results thus obtained were tabulated using customized perl scripts where

the cut-off E-value was set to 0.001. The final results obtained from design-island were fed into another perl program to generate a circular map of the chromosome indicating the putative GIs PAK6 as identified by design-island in separate phases using different colors. The algorithm is described in Fig. 1. Coordinates of statistically significant genomic segments of three V. cholerae strains under study were determined by design-island from two separate phases. From these predicted regions of three V. cholerae strains the coding regions were marked out with the protein table as the reference available at the NCBI database using a customized perl script. The results show that among the three strains under study, the maximum coverage by the GIs after the refinement phase was found to be 50.90% in the case of V. cholerae MJ1236 (large chromosome) while the least coverage was 33.11%, as in case of V. cholerae El Tor N16961 (small chromosome) as evident from Table 1. design-island identified all the known GIs of V.

For this the application weblogo 3 (Crooks et al., 2004; http://weblogo.threeplusone.com) was used. Sequence logos have been useful in visualizing patterns in aligned sequence motifs (Schneider & Stephens, 1990) and have indeed been used to analyse Tat motifs (see e.g. Bendtsen et al., 2005). We used this to compare the Tat motifs of haloarchaeal Tat substrates with that of the consensus E. coli GSI-IX chemical structure motif

(S/TRRxFLK). Signal peptide-containing sequences were extracted from genomes of E. coli and three fully sequenced haloarchaea: H. marismortui, N. pharaonis, and H. salinarum. The datasets obtained (see Supporting Information, Table S1) were filtered as outlined in Materials and methods to minimize the number of false-positive hits. Current information of prokaryotic signal peptides in general and the Tat system more specifically is mostly derived from bacterial systems, and as such, our searches may have been biased towards bacterial-like signal peptides. This, and our additional filtering, has most likely led to the absence of some genuine Tat signal peptides. Indeed, some proteins that are known to be Tat substrates in E. coli are missing from our dataset, including FdnH, HyaA, and HybO, all of which have been shown experimentally selleck to be Tat substrates (Hatzixanthis et al., 2003; Berks et al., 2005).

However, these three contain C-terminal transmembrane helices, which is the reason why our filtering steps rejected them. Nevertheless, only a fairly small proportion of Tat substrates have such additional

membrane-spanning domains, and we think that this approach has also resulted in datasets with very few or no false-positive proteins. The twin-arginine motifs obtained were aligned manually and used to generate sequence logos (Fig. 1). As can be observed from the top panel, our method used indeed led to a motif with the consensus SRRxFLK over as observed before (Berks, 1996). The twin-arginine motifs in haloarchaea were similar, but with a number of notable differences. Firstly, the dominance of Phe in position 5 is less pronounced than in E. coli; Val is found in that position in a very similar frequency. Secondly, Leu in position 6 appears to be far more frequent in haloarchaeal Tat motifs as compared with the E. coli Tat motif. Finally, the Lys in position 7 is less common in haloarchaea as compared with E. coli. Some of these differences may be attributable to the overall differences in the amino acid composition between halophilic and nonhalophilic proteins. For instance, haloarchaea contain, on average, fewer large hydrophobic residues such as Phe, as well as a relatively low percentage of lysine residues as compared with bacteria such as E. coli or Bacillus subtilis (Bolhuis et al., 2007). In this respect, the prominence of Leu in position 6 is actually interesting as this residue is, like Phe, less frequent in haloarchaeal proteins.

, 2009). We also found evidence of genetic exchange between Xanthomonas and Betaproteobacteria. A contig from Xcm 4381 (Fig. 2c) most

closely resembled the genome of Acidovorax species JS42 (95% sequence identity over 7935 nucleotides) and, slightly more distantly (94% identity over 3327 nucleotides), resembled the genome of X. campestris pathovar vesicatoria 85-10. This region encodes a predicted selleck screening library TrbK-like protein. TrbK is usually plasmid associated (Haase et al., 1996), but the corresponding genomic regions in Acidovorax species JS42 and in X. campestris pathovar vesicatoria 85-10 appear to be chromosomally located. It is unclear whether the 23-kb Xcm 4381 contig (Fig. 2c) represents a plasmid or is part of the chromosome. Plant-pathogenic Xanthomonas pathovars require a T3SS to secrete and translocate effector proteins (Alfano & Collmer, 2004; Yang et al., 2005; Grant et al., 2006; Gurlebeck et al., 2006;

White et al., 2006, 2009; Kay & Bonas, 2009; Buttner & Bonas, 2010) in order to cause disease. These effectors have evolved to manipulate host cellular processes to the benefit of the pathogen; however, many plants have evolved resistance whereby they can recognize specific effectors, triggering the hypersensitive response. Therefore, in the context of a resistant plant, these effectors show an ‘avirulence’ activity, thus limiting the pathogen’s host range (Alfano & Collmer, 2004; Yang et al., 2005; Grant et al., 2006; Gurlebeck et al., 2006; White et al., 2006, 2009; ifoxetine Kay & Bonas, 2009; Buttner & 3-MA manufacturer Bonas, 2010). A single Xanthomonas genome

typically encodes 20–30 T3SS effectors. The repertoire of effectors varies between species and strains within species and is believed to be a key determinant in the host range of a given pathogen. The draft genomes of both Xcm 4381 and Xvv 702 encoded a complete T3SS apparatus. To identify homologues of known T3SS effectors, we used blast searches against catalogues of proteins from the Pseudomonas syringae Hop Identification and Nomenclature Home Page (http://www.pseudomonas-syringae.org/), The Xanthomonas Resource (http://www.xanthomonas.org/t3e.html) and papers by White et al. (2009) and Gurlebeck et al. (2006). In common with all previously sequenced Xanthomonas genomes, both draft genomes encode homologues of the candidate T3SS effectors AvrBs2, AvrGf1, XopF, XopK, XopL, XopN, XopP, XopQ, XopR, XopX and XopZ. Both strains also encode homologues of XopA, XopB, XopG, XopH, XopI, XopY, XopAA, XopAD, XopAE and XopAK, which are conserved in a subset of the previously sequenced Xanthomonas genomes (http://www.xanthomonas.org/t3e.html). Both Xcm 4381 and Xvv 702 also encode proteins sharing 71% amino acid sequence identity with P. syringae effector HopW1; these have no significant sequence similarity to any known Xanthomonas protein (Fig. 3). Both draft genomes contained genes encoding homologues of the P.

2,5,50 The extent of this risk is not well understood or easily predicted. Some individuals have demonstrated the

ability to function well at high altitude whereas others suffer the consequences of increased pulmonary hypertension, HAPE, or right heart failure even at moderate altitudes.50–56 Symptoms with ascent may include dyspnea, weakness on exertion, and syncope.5 For people with symptomatic pulmonary hypertension at sea level, altitude exposure is contraindicated.2 click here Asymptomatic patients with CHD should be warned of the potential for developing HAPE and take nifedipine prophylactically to reduce their risk. Travelers with a brisk hypoxic pulmonary vasoconstrictor response may be identified in the clinic by observing their response to inhalation of a low oxygen mixture.5 These recommendations equally apply to patients with primary or secondary pulmonary hypertension.5 People with chronic obstructive pulmonary disease (COPD) may be hypoxemic at sea level and thus may develop altitude-related BIRB 796 cost symptoms at lower elevations than healthy people (Figure 2).2,8,27 Blunted carotid body

response due to chronic hypercapnia may reduce their ability to produce a hypoxic ventilatory response, thus further exacerbating the hypoxia.7 Breathing cold air results in pulmonary vasoconstriction and increased pulmonary artery pressure.8,57 Elevated levels of carboxyhemoglobin due to smoking may further compromise oxygen-carrying capacity in this cohort.58 Depending on baseline oxygen saturation and the PAK6 pathological condition

of the lungs, risks associated with altitude exposure include profound hypoxemia, pulmonary hypertension, disordered ventilatory control, impaired respiratory muscle function, and sleep-disordered breathing.2 No studies have been conducted on patients with COPD at high altitude. However, studies of patients with mild to moderate COPD at 1,920 m concluded that it is safe for such patients to travel to intermediate altitude.33,58 Altitude exposure is contraindicated for patients with severe COPD who have dyspnea at rest or on mild exertion at sea level. Patients with moderate disease should undergo individualized risk assessment and ascend with caution.2,7 Hypoxic challenge, spirometry testing, and the British Thoracic Society’s (BTS)59 guidelines for respiratory patients planning air travel may provide useful guidance for physicians.2,7,27 To minimize the risk of adverse effects, patients with COPD should avoid strenuous exercise at altitude and ensure optimal health prior to ascent.27 Maintenance of hydration at altitude is important to avoid problems associated with thickened mucosal secretions.60 Altitude can influence bronchial hyperresponsiveness, and thus, the likelihood of an acute asthma attack.

A 1 g L−1 stock was prepared in acetone and exposure solutions were made from it. All other chemicals were of analytical selleck chemical grade procured from local commercial

sources. Gram-negative representative strain E. coli K12 and Gram-positive representative strain B. subtilis B19 were kept in the laboratory of Environment Microbiology and Microbial Molecular Ecology, Northeast Agricultural University. Escherichia coli K12 and B. subtilis B19 were inoculated into LB medium (composed of 10 g tryptone, 5 g yeast extract, 10 g NaCl, 1000 mL H2O, adjusted pH to 7.0 and autoclaved to sterilize) and incubated at 30 °C with shaking at 130 r.p.m. To elucidate the toxicity and influence of atrazine to bacteria, the bacterial growth rate in the presence and absence of 500 μg L−1 atrazine in LB medium were investigated by reading the optical density at a wavelength of 600 nm (OD600 nm) every 4 h. Escherichia coli K12 and B. subtilis B19 were incubated in LB AZD1208 cell line medium for 24 h. Atrazine was

added to final concentrations of 0, 100, 200, 500, 800 and 1000 μg L−1. Bacterial cells were harvested after treatment with atrazine for 0, 6, 12 and 24 h. Bacterial cells (20 mL) harvested from liquid LB medium were centrifuged at 10 000 g for 10 min, washed twice with ice-cold 0.9% sodium chloride solution, and resuspended in a fresh 0.9% sodium chloride solution (5 mL) prior to lysis. Cell suspensions were then subjected to 99 rounds of sonication in an ice-water bath for 3 s, followed by cooling for

another 3 s. The debris was removed by centrifugation at 15 000 g for 10 min. The supernatant was HSP90 transferred to a new sterile centrifugal tube and used for enzyme activity assay directly. Protein concentration, SOD, CAT, GST activities and T-AOC were determined spectrophotometrically at 595, 550, 240, 412 and 520 nm, respectively, using commercial kits A045, A001, A007, A004 and A015 (Nanjing Jiancheng Bioengineering Institute, Jiangsu Province, China) (Lü et al., 2009). One unit of SOD was defined as the amount of the enzyme which gave 50% inhibition of the oxidation rate of 0.1 mM pyrogallol in 1 mL of solution at 25 °C (Zhang et al., 2005). One unit of CAT was defined as the amount of lysate that decomposed 1 μmol of H2O2 at pH 7.0 and 25 °C in 1 min (Lü et al., 2009). One unit of GST was defined as the amount of lysate that decomposed 1.0 μM of GSH at 37 °C in 1 min, excluding non-enzymatic reaction. One unit of T-AOC was defined as the increment in the absorbance by 0.01 at 37 °C in 1 min. The protein concentration in cell lysates was determined by a modified Lowry procedure using bovine serum albumin as the standard. The specific SOD and CAT activities were expressed as U mg protein–1. Data were expressed as means ± SE of six replicates from two independent experiments. Data were analyzed by one-way analysis of variance. Mean values were compared by Duncan’s new multiple range test at the 5% level using spss 17.0 software.

The cases were classified following the EORTC/MSG Consensus Group criteria (European Organization for Research and Treatment of Cancer/Mycosis Study Group).27 Proven histoplasmosis or PCM was considered when the fungus was recovered in culture from a specimen or when the microorganism was observed using histopathology or direct microscopy. Cases were classified as probable when a consistent clinical picture was found and we had a positive result in an immunodiffusion test (ID Fungal Antibody System, Immuno-Mycologics Inc, Norman, OK, USA). All cases except one fulfilled the proven

or probable criterion. In Patient 11, there were only clinical suspicion and positive results by RT-PCR (Table 2). This case was classified as possible. Apitolisib solubility dmso Using epidemiological criteria, we also classified the cases as either travelers or immigrants and people who had lived in an endemic region for a long period of time. In case of H capsulatum strains, mycelia were stained with lactophenol cotton blue dye (Difco, Soria-Melguizo, Madrid, Spain). Characteristic macroconidia were observed microscopically.

Extraction of nucleic acids from clinical strains was undertaken in Biosafety Level III facilities and in compliance with Spanish Laws (Royal Decree 664/1997). DNA extraction from strains and clinical samples was performed as described by Buitrago and colleagues.20 DNA extracted from clinical strains was used to amplify the internal transcriber spacer (ITS) region.28 Sequence analysis of amplified fragments was performed by

comparing the DNA sequences with the ITS sequences of H capsulatum selleck chemical var. duboisii (ATCC 24295), H capsulatum var. capsulatum (CBS207.55 and CBS214.53), and P brasiliensis (ATCC32069 and ATCC60855) obtained from the GenBank database (http://www.ncbi.nlm.nih.gov/Genbank/). RT-PCR for the detection of H capsulatum was carried out following the protocol described by Buitrago and colleagues.19 Primers and probes were designed on the basis of the nucleotide sequence of the ITS1 rDNA region. Probes were marked using fluoresce resonance energy transfer (FRET) technology, and the PCR NADPH-cytochrome-c2 reductase reactions were performed in the Lightcycler 480 (Roche Applied Science, Madrid, Spain). An internal control was included in the RT-PCR reaction following the Brugraff method.20,29 RT-PCR for the detection of P brasiliensis DNA was performed as described by Buitrago and colleagues.25 Detection of precipitating antibodies in patient’s sera was performed by an immunodiffusion test following manufacturer’s recommendations (ID Fungal Antibody System). This commercial test uses the antigens M and H against histoplasmosis sera and antigen gp43 against PCM sera. A total of 39 cases of histoplasmosis and 6 cases of PCM have been diagnosed in the Spanish Mycology Reference Laboratory in the last 3 years.

Each series PLX3397 datasheet contained at least 40 sections. Sections were viewed and analyzed with a light Axio Imager A1 (Carl Zeiss) microscope; images were captured using an AxioCam (Carl Zeiss) digital camera with the software axiovision ac. Trace software igl trace 1.26b (Fiala, 2002) was used

to adjust serial sections by their contours. The 3D images were exported to WRML format, with final rendering using 3D Studio Max9 (Autodesk). Electron microscopy of the cell surfaces of yeasts grown in oil-containing media revealed profound structural alterations in the cell walls as compared with yeasts grown without hydrocarbons. Depending on the species, the yeasts could be divided into two groups. Group one included several species of Candida, Torulopsis and Shwanniomyces. When grown on either hexadecane, n-alkane mixtures (C12–C20) or crude oil, these yeasts formed ‘canals’ in their cell walls. The canals were numerous, with up to 100 canals per cell, and were especially vivid on the carbon–platinum replicas of the cell surface (Fig. 1a and b). The formation of the canals was substrate-dependent. When cells that had been grown on oil were transferred

to a medium with glucose as a sole carbon source, the canal-forming yeasts reverted to a morphotype without canals (Fig. 1c). Along with canal formation, the yeasts S. occidentalis, T. candida and C. maltosa also secreted copious amounts of fibrilar substances when cultivated in media with hexadecane, a mixture of n-alkanes or crude oil. Ultrathin sections and freeze fraction micrographs vividly Carfilzomib demonstrated that this exosubstance was anatomically bound with the canal features (Fig. 2a–c). Cytochemical staining Farnesyltransferase of these cells with diaminobenzidine further revealed the presence of oxidative enzymes

that were concentrated in the canals (Fig. 3a and b). The oxidative enzymes could also be observed in canals of partially purified cell wall fractions from these yeasts, suggesting that these enzymes are ionically or covalently bound with these modified sites of the cell wall (Fig. 3c). Immunocytochemical staining (Fig. 3d) further showed that cytochrome P-450 was concentrated in distinct locations within the cell walls. All of these facts confirmed the supposition that primary oxidation of hydrocarbons by yeasts occurs mainly in the canals where degradative enzymes are entrapped in a polymer matrix. In contrast to the canal-forming yeast, a second group of yeasts including C. lipolytica and C. paralipolytica did not form canals when grown in hydrocarbon- or crude oil-containing media, but instead secreted large amounts of fibrilar substances. The carbon–platinum replicas of these yeasts were quite smooth (Fig. 3e) and the yeasts appeared to secrete the fibrilar substances over their entire cell surface. The products of the cytochemical staining reaction targeting oxidative enzymes were located both on the cell surfaces and on the exocellular films (Fig. 3f).

lactis ssp. cremoris SMBI198, a strain derived from NZ9000, knocked out in the chromosomal htrA gene (Poquet et al., 2000; Rigoulay et al., 2004). The resulting strain produced only a surface-associated see more form of the recombinant flagellin (Fig. 1b). Interestingly, two bands showed homology with B. cereus flagellin, one of around 45 and the other one of around 63 kDa. It is known that, in certain cases, protein aggregates are difficult to disassociate, producing this kind of artefact

in SDS-PAGE (Kankainen et al., 2009). This tendency to aggregation may lead to bacterial autoaggregation, and the physical–chemical dynamics of this process are currently under investigation. This could reflect the trend to auto-assembly that flagellins display in vivo (Hueck, 1998). In addition, it reinforces the role of HtrA as the major housekeeping protease on the L. lactis surface as, in the absence of it, the aggregated flagellin cannot be proteolyzed and thus shed into the bacterial surroundings. Lactococcuslactis ssp. cremoris CH showed a better ability to adhere to mucin when flagellin production was induced with nisin (Fig. 2). Adhesion of both

L. lactis ssp. cremoris SMBI198 and L. lactis ssp. cremoris SMBI198 (pNZ8110) strains was similar to uninduced L. lactis ssp. cremoris CH cultures (data not shown). After gene induction, the adhesion was increased by a factor CHIR-99021 manufacturer of 4.7. Nisin-induced L. lactis CH cultures inhibited the adhesion to mucin of the two enteropathogens used in this study in a dose-dependent manner (Fig. 2). A lower inhibition was also observed when uninduced L. lactis ssp. cremoris

CH cultures were used. This is not surprising as L. lactis is also able to bind to mucin; an interference with enteropathogen adhesion to mucin is thus expected. The adhesion data, corrected by the inhibitory effect observed for the uninduced L. lactis CH strain cultures, showed that L. lactis ssp. cremoris CH expressing the Bacillus flagellin was able to inhibit the adhesion of E. coli 4.4 times (1.8 times in the case of uninduced cultures), and 3.9 times in the case of S. enterica (1.6 times in the case of uninduced cultures), when the E. coli/L. lactis ratio was 1 : 10. In our previous work, we showed that adhesion of B. cereus CH to mucin could be explained, Dichloromethane dehalogenase to a large extent, by the presence of a flagellin on its surface (Sánchez et al., 2009a). Flagella have been proposed as important factors for bacterial adhesion to mucosal surfaces (Rumbo et al., 2006), and are glycosylated in several microorganisms (Ewing et al., 2009; Hayakawa et al., 2009; Konishi et al., 2009; Logan et al., 2009). One B. cereus strain lacking the flhA gene, which results in the absence of flagella, presented a lower adhesion to both Caco-2 and HeLa cell lines (Ramarao & Lereclus, 2006). In addition, monomeric flagellins detached from the cell surface have been proposed as the soluble probiotic factor secreted by the strain E. coli Nissle 1917.

In the ΔAoatg15 mutant, autophagic bodies accumulated in vacuoles, Small molecule high throughput screening suggesting that the uptake process proceeded. We therefore propose that the level of autophagy is closely correlated with the degree of differentiation in A. oryzae. In eukaryotes, macroautophagy (autophagy) is a conserved degradation process that mediates the trafficking of cytosolic proteins and organelles into lysosomes/vacuoles for bulk degradation (Reggiori & Klionsky, 2002). Although the process appears to predominantly recycle

macromolecules and aid cell survival during periods of nutritional starvation, autophagy is also involved in development and differentiation in numerous eukaryotes, including yeasts, plants, and

mammals, among others (Levine & Klionsky, 2004). This involvement may have resulted from the autophagic degradation of damaged organelles and cytosol for constitutive cell clearance and cellular remodeling during development and differentiation. The autophagic process proceeds sequentially through several steps, involving the induction of autophagy, formation of autophagosomes, fusion of autophagosomes to lysosomes/vacuoles, and degradation of autophagic bodies selleck chemicals llc (Mizushima, 2007; Pollack et al., 2009). In Saccharomyces cerevisiae, the induction of autophagy results from inactivation of the target of rapamycin (Tor) kinase, allowing formation of the Atg1 kinase complex composed of Atg1, Atg13, and Atg17 (Funakoshi et al., 1997; Kamada et al., 2000; Kabeya et al., 2005). The association of Atg13 with Atg1, which is essential for autophagy, is prevented by phosphorylation of Atg13 in a Tor kinase-dependent manner under conditions suitable for growth. In starvation conditions, Atg13 is dephosphorylated by inhibition of Tor kinase activity, allowing it to associate with Atg1 (Kamada 5-Fluoracil supplier et al., 2000). The induction of autophagy induces the formation of cup-shaped isolation membranes, which subsequently

elongate and sequester cytosol and/or organelles within double-membrane vesicles termed autophagosomes. Saccharomyces cerevisiae Atg8 is a ubiquitin-like protein that is essential for the formation of autophagosomes and is localized in preautophagosomal structures (PAS) and the membranes of autophagosomes and autophagic bodies, and has been used as a marker for these organelles (Suzuki et al., 2001). A critical event for autophagy involves the conjugation of the carboxy (C)-terminal glycine of Atg8 with phosphatidylethanolamine (PE), which is mediated by a ubiquitination-like system composed of Atg4 (cysteine protease), Atg7 (E1-like protein), and Atg3 (E2-like protein) (Ichimura et al., 2000; Kirisako et al., 2000). Atg4 cleaves newly synthesized Atg8 to expose the C-terminal glycine for conjugation with PE, and also cleaves Atg8-conjugated PE (Atg8-PE) to recycle Atg8.

3.1[12] prescribed and dispensed in Wales were extracted from CASPA.net for the period June 2004 to December 2010 (12 months before and 66 months after OTC ophthalmic chloramphenicol availability). OTC sales data were obtained from IMS Health and included four established proprietary brands of both chloramphenicol eye drops and ointment (Brochlor, Golden Eye Antibiotic, Galpharm Vision, Optrex Infected Eyes), together with one proprietary brand (Tubilux) and one own-brand of eye drops. As at December

2010, there were two further proprietary brands of chloramphenicol eye drops available as P medicines in the UK[25] but data for these products were unavailable and thus not included in the analysis. Ophthalmic chloramphenicol preparations licensed as POMs, such as Minims eye drops, were excluded from the OTC sales analysis. The OTC sales

data obtained were available from June 2005 to December 2010 (66 months) and Selleckchem SCH772984 represented the supply of ophthalmic chloramphenicol preparations from wholesalers into 614/708 (87%) NHS-contracted community pharmacies in Wales. Data for the remaining 94 NHS-contracted pharmacies and eight pharmacies without NHS contract selleck compound were obtained direct from the pharmacy chain concerned (Company A) for the period January 2008 to December 2010 (36 months). OTC sales of chloramphenicol eye drops from Company A between June 2005 and December 2007 (30 months) and ointment between July and December 2007 (6 months) were estimated using linear regression. The line of best fit generated from the model was extrapolated backwards based on available cumulative sales data. The OTC sales from Company A (estimated and actual) were combined with IMS Health sales data to give the total quantity of OTC ophthalmic chloramphenicol sold in Wales from June 2005 to December 2010. The total number of items supplied on prescription or sold OTC are presented as the 12 month totals for the eye drops, from June to May, and for the ointment, from

July to June, to allow the comparison before and after their respective availability OTC. The correlation coefficient (r) for prescription items supplied and OTC sales of combined chloramphenicol eye drops and ointment was calculated Flavopiridol (Alvocidib) using Spearman’s rank correlation, based on actual prescribing and OTC sales data between January 2008 and December 2010. All data analysis and statistics were performed using PASW version 18 (SPSS, Chicago, IL, USA). The linear regression model generated cumulative sales equations for eye drops (R2 = 0.998, P < 0.0001) and eye ointment (R2 = 0.995, P < 0.0001) for Company A and estimated cumulative sales for the respective periods when no data were available (data not shown). The total cumulative quantities of ophthalmic chloramphenicol sold OTC (IMS Health + Company A; actual and estimated OTC sales) are shown in Figure 1. The supply of chloramphenicol eye drops from 2004–2005 to 2009–2010 is shown in Figure 2.