Optimising aerial military medical evacuation dispatching decisions via operations research techniques.

Senior military leaders and medical practitioners continuously seek new ways to improve the performance and organisation of deployed medical evacuation (MEDEVAC) systems to minimise mortality rates of combat casualties. The objective of this paper is to highlight how recent research in the fields of operations research and machine learning can be leveraged to better inform the implementation and modification of current and future MEDEVAC tactics, techniques and procedures for combat operations in a deployed environment. More specifically, this paper discusses state-of-the-art techniques that optimise the management of MEDEVAC assets prior to and during combat operations. These recent research efforts emphasise that military healthcare administrators should contribute to and extend the evolving portfolio of research that seeks to design and develop decision support systems leveraging artificial intelligence and operations research to improve MEDEVAC system performance.

Solidification fronts in supercooled liquids: how rapid fronts can lead to disordered glassy solids.

We determine the speed of a crystallization (or, more generally, a solidification) front as it advances into the uniform liquid phase after the system has been quenched into the crystalline region of the phase diagram. We calculate the front speed by assuming a dynamical density functional theory (DDFT) model for the system and applying a marginal stability criterion. Our results also apply to phase field crystal (PFC) models of solidification. As the solidification front advances into the unstable liquid phase, the density profile behind the advancing front develops density modulations and the wavelength of these modulations is a dynamically chosen quantity. For shallow quenches, the selected wavelength is precisely that of the crystalline phase and so well-ordered crystalline states are formed. However, when the system is deeply quenched, we find that this wavelength can be quite different from that of the crystal, so the solidification front naturally generates disorder in the system. Significant rearrangement and aging must subsequently occur for the system to form the regular well-ordered crystal that corresponds to the free energy minimum. Additional disorder is introduced whenever a front develops from random initial conditions. We illustrate these findings with simulation results obtained using the PFC model.

Modeling the structure of liquids and crystals using one- and two-component modified phase-field crystal models.

A modified phase-field crystal model in which the free energy may be minimized by an order parameter profile having isolated bumps is investigated. The phase diagram is calculated in one and two dimensions and we locate the regions where modulated and uniform phases are formed and also regions where localized states are formed. We investigate the effectiveness of the phase-field crystal model for describing fluids and crystals with defects. We further consider a two-component model and elucidate how the structure transforms from hexagonal crystalline ordering to square ordering as the concentration changes. Our conclusion contains a discussion of possible interpretations of the order parameter field.

Modelling the evaporation of thin films of colloidal suspensions using dynamical density functional theory.

Recent experiments have shown that various structures may be formed during the evaporative dewetting of thin films of colloidal suspensions. Nanoparticle deposits of strongly branched 'flower-like', labyrinthine and network structures are observed. They are caused by the different transport processes and the rich phase behaviour of the system. We develop a model for the system, based on a dynamical density functional theory, which reproduces these structures. The model is employed to determine the influences of the solvent evaporation and of the diffusion of the colloidal particles and of the liquid over the surface. Finally, we investigate the conditions needed for 'liquid-particle' phase separation to occur and discuss its effect on the self-organized nanostructures.

Coronary artery disease in a large renal transplant population: implications for management.

Coronary artery disease (CAD) accounts for approximately one-half of the sizable mortality in patients with end-stage renal disease who have undergone transplantation. The study was a retrospective review of 1460 patients who underwent renal transplantation at the Mount Sinai Medical Center from January 1, 2000 to October 31, 2009. Noninvasive stress testing was performed in 848 patients (88.1%) with 278 patients (32.8%) having abnormal results. Cardiac catheterization was performed in 357 patients (37.1%) and of these, 212 patients had obstructive disease (59.4%). At 5 years posttransplant, there was no statistically significant difference between those with nonobstructive CAD and those who required percutaneous or surgical interventions (adjusted hazard ratio [aHR], 1.243; CI 95%, 0.513-3.010; p = 0.630). Those with medically managed obstructive CAD had significantly higher rates of death at the 5-year period when compared to those who received percutaneous intervention (aHR, 3.792; CI 95%, 1.320-10.895; p = 0.013) or those who received coronary artery bypass grafting (aHR, 6.691; CI 95%, 1.200-37.323). Because noninvasive imaging is poorly predictive of coronary disease in this high-risk population, an anatomic diagnosis is recommended. Revascularization may result in improved long-term outcomes.

Dynamical density functional theory for the dewetting of evaporating thin films of nanoparticle suspensions exhibiting pattern formation.

Recent experiments have shown that the striking structure formation in dewetting films of evaporating colloidal nanoparticle suspensions occurs in an ultrathin "postcursor" layer that is left behind by a mesoscopic dewetting front. Various phase change and transport processes occur in the postcursor layer that may lead to nanoparticle deposits in the form of labyrinthine, network, or strongly branched "finger" structures. We develop a versatile dynamical density functional theory to model this system which captures all these structures and may be employed to investigate the influence of evaporation or condensation, nanoparticle transport, and solute transport in a differentiated way. We highlight, in particular, the influence of the subtle interplay of decomposition in the layer and contact line motion on the observed particle-induced transverse instability of the dewetting front.

Modelling approaches to the dewetting of evaporating thin films of nanoparticle suspensions.

We review recent experiments on dewetting thin films of evaporating colloidal nanoparticle suspensions (nanofluids) and discuss several theoretical approaches to describe the ongoing processes including coupled transport and phase changes. These approaches range from microscopic discrete stochastic theories to mesoscopic continuous deterministic descriptions. In particular, we describe (i) a microscopic kinetic Monte Carlo model, (ii) a dynamical density functional theory and (iii) a hydrodynamic thin film model. Models (i) and (ii) are employed to discuss the formation of polygonal networks, spinodal and branched structures resulting from the dewetting of an ultrathin 'postcursor film' that remains behind a mesoscopic dewetting front. We highlight, in particular, the presence of a transverse instability in the evaporative dewetting front, which results in highly branched fingering structures. The subtle interplay of decomposition in the film and contact line motion is discussed. Finally, we discuss a simple thin film model (iii) of the hydrodynamics on the mesoscale. We employ coupled evolution equations for the film thickness profile and mean particle concentration. The model is used to discuss the self-pinning and depinning of a contact line related to the 'coffee-stain' effect. In the course of the review we discuss the advantages and limitations of the different theories, as well as possible future developments and extensions.

Differential expression of IEG mRNA in rat brain following acute treatment with clozapine or haloperidol: a semi-quantitative RT-PCR study.

Antipsychotic drugs have been shown to modulate immediate early gene (IEG) expression in rat brain regions that are associated with schizophrenia, which may be directly linked to their immediate therapeutic benefit. In this study, we analysed the expression profile of a series of IEGs (c-fos, c-jun, fra-1, Krox-20, Krox-24, arc, sgk-1, BDNF and NARP) in six rat brain regions (prefrontal cortex, hippocampus, striatum, nucleus accumbens, thalamus and cerebellum). Rats (n=5) were administered either clozapine (20 mg/kg i.p.), haloperidol (1 mg/kg i.p.) or the appropriate vehicle with pre-treatment times of 1, 6 and 24 h. IEG expression was analysed in these regions by Taqman RT-PCR. The spatial and temporal profile of IEG induction following antipsychotic drug treatment correlates with regions associated with the efficacy and side effect profile of each drug. In particular, sgk-1 expression levels after antipsychotic drug treatment may have predictive value when investigating the profile of a novel antipsychotic drug.

Molecular cloning and characterisation of a novel GABAB-related G-protein coupled receptor.

Using a homology-based bioinformatics approach we have analysed human genomic sequence and identified the human and rodent orthologues of a novel putative seven transmembrane G protein coupled receptor, termed GABA(BL). The amino acid sequence homology of these cDNAs compared to GABA(B1) and GABA(B2) led us to postulate that GABA(BL) was a putative novel GABA(B) receptor subunit. The C-terminal sequence of GABA(BL) contained a putative coiled-coil domain, di-leucine and several RXR(R) ER retention motifs, all of which have been shown to be critical in GABA(B) receptor subunit function. In addition, the distribution of GABA(BL) in the central nervous system was reminiscent of that of the other known GABA(B) subunits. However, we were unable to detect receptor function in response to any GABA(B) ligands when GABA(BL) was expressed in isolation or in the presence of either GABA(B1) or GABA(B2). Therefore, if GABA(BL) is indeed a GABA(B) receptor subunit, its partner is a potentially novel receptor subunit or chaperone protein which has yet to be identified.

Comparative immunohistochemical localisation of GABA(B1a), GABA(B1b) and GABA(B2) subunits in rat brain, spinal cord and dorsal root ganglion.

GABA(B) receptors are G-protein-coupled receptors mediating the slow onset and prolonged synaptic actions of GABA in the CNS. The recent cloning of two genes, GABA(B1) and GABA(B2), has revealed a novel requirement for GABA(B) receptor signalling. Studies have demonstrated that the two receptor subunits associate as a GABA(B1)/GABA(B2) heterodimer to form a functional GABA(B) receptor. In this study we have developed polyclonal antisera specific to two splice variants of the GABA(B1) subunit, GABA(B1a) and GABA(B1b), as well as an antiserum to the GABA(B2) subunit. Using affinity-purified antibodies derived from these antisera we have mapped out the distribution profile of each subunit in rat brain, spinal cord and dorsal root ganglion. In brain the highest areas of GABA(B1a), GABA(B1b) and GABA(B2) subunit expression were found in neocortex, hippocampus, thalamus, cerebellum and habenula. In spinal cord, GABA(B1) and GABA(B2) subunits were expressed in the superficial layers of the dorsal horn, as well as in motor neurones in the deeper layers of the ventral horn. GABA(B) receptor subunit immunoreactivity in dorsal root ganglion suggested that expression of GABA(B1b) was restricted to the large diameter neurones, in contrast to GABA(B1a) and GABA(B2) subunits which were expressed in both large and small diameter neurones. Although expression levels of GABA(B1) and GABA(B2) subunits varied we found no areas in which GABA(B1) was expressed in the absence of GABA(B2). This suggests that most, if not all, GABA(B1) immunoreactivity may represent functional GABA(B) receptors. Although our data are in general agreement with functional studies, some discrepancies in GABA(B1) subunit expression occurred with respect to other immunohistochemical studies. Overall our data suggest that GABA(B) receptors are widely expressed throughout the brain and spinal cord, and that GABA(B1a) and GABA(B1b) subunits can associate with GABA(B2) to form both pre- and post-synaptic receptors.

GABA(B2) is essential for g-protein coupling of the GABA(B) receptor heterodimer.

GABA(B) receptors are unique among G-protein-coupled receptors (GPCRs) in their requirement for heterodimerization between two homologous subunits, GABA(B1) and GABA(B2), for functional expression. Whereas GABA(B1) is capable of binding receptor agonists and antagonists, the role of each GABA(B) subunit in receptor signaling is unknown. Here we identified amino acid residues within the second intracellular domain of GABA(B2) that are critical for the coupling of GABA(B) receptor heterodimers to their downstream effector systems. Our results provide strong evidence for a functional role of the GABA(B2) subunit in G-protein coupling of the GABA(B) receptor heterodimer. In addition, they provide evidence for a novel "sequential" GPCR signaling mechanism in which ligand binding to one heterodimer subunit can induce signal transduction through the second partner of a heteromeric complex.

The C-terminal domains of the GABA(b) receptor subunits mediate intracellular trafficking but are not required for receptor signaling.

GABA(B) receptors are G-protein-coupled receptors that mediate slow synaptic inhibition in the brain and spinal cord. These receptors are heterodimers assembled from GABA(B1) and GABA(B2) subunits, neither of which is capable of producing functional GABA(B) receptors on homomeric expression. GABA(B1,) although able to bind GABA, is retained within the endoplasmic reticulum (ER) when expressed alone. In contrast, GABA(B2) is able to access the cell surface when expressed alone but does not couple efficiently to the appropriate effector systems or produce any detectable GABA-binding sites. In the present study, we have constructed chimeric and truncated GABA(B1) and GABA(B2) subunits to explore further GABA(B) receptor signaling and assembly. Removal of the entire C-terminal intracellular domain of GABA(B1) results in plasma membrane expression without the production of a functional GABA(B) receptor. However, coexpression of this truncated GABA(B1) subunit with either GABA(B2) or a truncated GABA(B2) subunit in which the C terminal has also been removed is capable of functional signaling via G-proteins. In contrast, transferring the entire C-terminal tail of GABA(B1) to GABA(B2) leads to the ER retention of the GABA(B2) subunit when expressed alone. These results indicate that the C terminal of GABA(B1) mediates the ER retention of this protein and that neither of the C-terminal tails of GABA(B1) or GABA(B2) is an absolute requirement for functional coupling of heteromeric receptors. Furthermore although GABA(B1) is capable of producing GABA-binding sites, GABA(B2) is of central importance in the functional coupling of heteromeric GABA(B) receptors to G-proteins and the subsequent activation of effector systems.

Cloning and characterization of human NTT5 and v7-3: two orphan transporters of the Na+/Cl- -dependent neurotransmitter transporter gene family.

Orphan transporters form a growing subfamily of genes related by sequence similarity to the Na+/Cl- -dependent neurotransmitter superfamily. Using a combination of database similarity searching and cloning methods, we have identified and characterized two novel human orphan transporter genes, v7-3 and NTT5. Similar to other known orphan transporters, v7-3 and NTT5 contain 12 predicted transmembrane domains, intracellular N- and C-terminal domains, and large extracellular loops between transmembrane (TM) domains 3 and 4 and between TM domains 7 and 8. Residues within the extracellular loops are also predicted to contain sites for N-linked glycosylation. Human v7-3, the species orthologue of rat v7-3, contains an open reading frame (ORF) of 730 amino acids. Human NTT5 is a new member of the orphan transporter family and has an ORF of 736 amino acids. The amino acid sequences of human v7-3 and NTT5 are greater than 50% similar to other known orphan neurotransmitter transporters and also show sequence similarity to the human serotonin and dopamine transporters. Radiation hybrid mapping located the human v7-3 and NTT5 genes on chromosomes 12q21.3-q21.4 and 19q13.1-q13.4, respectively. Human mRNA distribution analysis by TaqMan reverse transcription-polymerase chain reaction showed that v7-3 mRNA is predominantly expressed in neuronal tissues, particularly amygdala, putamen, and corpus callosum, with low-level expression in peripheral tissues. In contrast, NTT5 mRNA was highly expressed in peripheral tissues, particularly in testis, pancreas, and prostate. Transient transfection with epitope-tagged transporter constructs demonstrated v7-3 to be expressed at the cell surface, whereas NTT5 was predominantly intracellular, suggestive of a vesicular location. Although the substrates transported by these transporters remain unknown, their specific but widespread distribution suggests that they may mediate distinct and important functions within the brain and the periphery.

The expression of GABA(B1) and GABA(B2) receptor subunits in the cNS differs from that in peripheral tissues.

GABA(B) receptors are G-protein-coupled receptors that mediate the slow and prolonged synaptic actions of GABA in the CNS via the modulation of ion channels. Unusually, GABA(B) receptors form functional heterodimers composed of GABA(B1) and GABA(B2) subunits. The GABA(B1) subunit is essential for ligand binding, whereas the GABA(B2) subunit is essential for functional expression of the receptor dimer at the cell surface. We have used real-time reverse transcriptase-polymerase chain reaction to analyse expression levels of these subunits, and their associated splice variants, in the CNS and peripheral tissues of human and rat. GABA(B1) subunit splice variants were expressed throughout the CNS and peripheral tissues, whereas surprisingly GABA(B2) subunit splice variants were neural specific. Using novel antisera specific to individual GABA(B) receptor subunits, we have confirmed these findings at the protein level. Analysis by immunoblotting demonstrated the presence of the GABA(B1) subunit, but not the GABA(B2) subunit, in uterus and spleen. Furthermore, we have shown the first immunocytochemical analysis of the GABA(B2) subunit in the brain and spinal cord using a GABA(B2)-specific antibody. We have, therefore, identified areas of non-overlap between GABA(B1) and GABA(B2) subunit expression in tissues known to contain functional GABA(B) receptors. Such areas are of interest as they may well contain novel GABA(B) receptor subunit isoforms, expression of which would enable the GABA(B1) subunit to reach the cell surface and form functional GABA(B) receptors.

Molecular cloning and characterization of two novel retinoic acid-inducible orphan G-protein-coupled receptors (GPRC5B and GPRC5C).

Using homology searching of public databases with a metabotropic glutamate receptor sequence from Caenorhabditis elegans, two novel protein sequences (named RAIG-2 (HGMW-approved symbol GPRC5B) and RAIG-3 (HGMW-approved symbol GPRC5C) were identified containing seven putative transmembrane domains characteristic of G-protein-coupled receptors (GPCRs). RAIG-2 and RAIG-3 encode open reading frames of 403 and 442 amino acid polypeptides, respectively, and show 58% similarity to the recently identified retinoic acid-inducible gene-1 (RAIG-1, HGMW-approved symbol RAI3). Analysis of the three protein sequences places them within the type 3 GPCR family, which includes metabotropic glutamate receptors, GABA(B) receptors, calcium-sensing receptors, and pheromone receptors. However, in contrast to other type 3 GPCRs, RAIG-1, RAIG-2, and RAIG-3 have only short N-terminal domains. RAIG-2 and RAIG-3 cDNA sequences were cloned into the mammalian expression vector pcDNA3 with c-myc or HA epitope tags inserted at their N-termini, respectively. Transient transfection experiments in HEK239T cells using these constructs demonstrated RAIG-2 and RAIG-3 expression at the cell surface. Distribution profiles of mRNA expression obtained by semiquantitative Taq-Man PCR analysis showed RAIG-2 to be predominantly expressed in human brain areas and RAIG-3 to be predominantly expressed in peripheral tissues. In addition, expression of RAIG-2 and RAIG-3 mRNA was increased following treatment with all-trans-retinoic acid in a manner similar to that previously described for RAIG-1. Finally, RAIG-2 was mapped to chromosome 16p12 (D16S405-D16S3045) and RAIG-3 to chromosome 17q25 (D17S1352-D17S785). These results suggest that RAIG-1, RAIG-2, and RAIG-3 represent a novel family of retinoic acid-inducible receptors, most closely related to the type 3 GPCR subfamily, and provide further evidence for a linkage between retinoic acid and G-protein-coupled receptor signal transduction pathways.

The use of TaqMan RT-PCR assays for semiquantitative analysis of gene expression in CNS tissues and disease models.

TaqMan reverse transcription polymerase chain reaction (RT-PCR) is a recently developed technique which allows the measurement of an accumulating PCR product in real time. In the present study we have validated the use of TaqMan RT-PCR for mRNA localisation studies in human and rat tissues, and for the investigation of gene expression changes in CNS animal models. In human brain, D(2) receptor mRNA was enriched in caudate nucleus and putamen, whilst in rat brain, highest levels of D(2) receptor mRNA expression were observed in striatum and nucleus accumbens, consistent with the known distribution of this receptor in basal ganglia. In a rat model of permanent middle cerebral artery occlusion (pMCAO), endogenous interleukin-1 receptor antagonist (IL-1ra) mRNA was upregulated over 30-fold at 24 h post-lesion in both striatum and cortex ipsilateral to artery occlusion. Brain-derived neurotrophic factor (BDNF) mRNA was transiently upregulated 3.7-fold at 3 h, but not at 24 h or 3 days after induction of cortical spreading depression (CSD) in rats. Our observations in these two animal models using TaqMan RT-PCR were consistent with previous reports using other techniques. In conclusion, TaqMan RT-PCR assays provide a rapid and reliable method for semi-quantitative analysis of gene expression in the nervous system.

Characterization of the dimerization of metabotropic glutamate receptors using an N-terminal truncation of mGluR1alpha.

The metabotropic glutamate receptor mGluR1alpha in membranes isolated both from rat brain and from cell lines transfected with cDNA coding for the receptor migrates as a disulphide-bonded dimer on sodium dodecyl sulphate-polyacrylamide gels. Dimerization of mGluR1alpha takes place in the endoplasmic reticulum because it is not prevented by exposing transfected human embryonic kidney (HEK) 293 cells to the drug brefeldin A, a drug that prevents egress of proteins from the endoplasmic reticulum. Dimerization was also not dependent on protein glycosylation as it was not prevented by treatment of the cells with tunicamycin. Using a mammalian expression vector containing the N-terminal domain of mGluR1alpha, truncated just before the first transmembrane domain (NT-mGluR1alpha), we show that the N-terminal domain is secreted as a soluble disulphide-bonded dimeric protein. In addition, the truncated N-terminal domain can form heterodimers with mGluR1alpha when both proteins are cotransfected into HEK 293 cells. However, mGluR1alpha and its splice variant mGluR1beta did not form heterodimers in doubly transfected HEK 293 cells. These results show that although the N-terminal domain of mGluR1alpha is sufficient for dimer formation, other domains in the molecule must regulate the process.

Interactions of the C terminus of metabotropic glutamate receptor type 1alpha with rat brain proteins: evidence for a direct interaction with tubulin.

Metabotropic glutamate receptors (mGluRs) are coupled to G protein second messenger pathways and modulate glutamate neurotransmission in the brain, where they are targeted to specific synaptic locations. As part of a strategy for defining the mechanisms for the specific targeting of mGluR1alpha, rat brain proteins which interact with the intracellular carboxy terminus of mGluR1alpha have been characterized, using affinity chromatography on a glutathione S-transferase fusion protein that contains the last 86 amino acids of mGluR1alpha. Three of the proteins specifically eluted from the affinity column yielded protein sequences, two of which were identified as glyceraldehyde-3-phosphate dehydrogenase and beta-tubulin; the other was an unknown protein. The identity of tubulin was confirmed by western immunoblotting. Using a solid-phase binding assay, the mGluR1alpha-tubulin interaction was shown to be direct, specific, and saturable with a KD of 2.3+/-0.4 microM. In addition, mGluR1alpha, but not mGluR2/3 or mGluR4, could be coimmunoprecipitated from solubilized brain extracts with tubulin using anti-beta-tubulin antibodies. However, mGluR1alpha could not be coimmunoprecipitated with the tubulin binding protein gephyrin, nor could it be coimmunoprecipitated with PSD95. Collectively these data demonstrate that the last 86 amino acids of the carboxyl-terminal tail of mGluR1alpha are sufficient to determine its interaction with tubulin and that there is an association of this receptor with tubulin in rat brain.

Interactions of the C Terminus of Metabotropic Glutamate Receptor Type 1α with Rat Brain Proteins: Evidence for a Direct Interaction with Tubulin.

Metabotropic glutamate receptors (mGluRs) are coupled to G protein second messenger pathways and modulate glutamate neurotransmission in the brain, where they are targeted to specific synaptic locations. As part of a strategy for defining the mechanisms for the specific targeting of mGluR1 α, rat brain proteins which interact with the intracellular carboxy terminus of mGluR1 α have been characterized, using affinity chromatography on a glutathione S-transferase fusion protein that contains the last 86 amino acids of mGluR1 α. Three of the proteins specifically eluted from the affinity column yielded protein sequences, two of which were identified as glyceraldehyde-3-phosphate dehydrogenase and ß-tubulin ; the other was an unknown protein. The identity of tubulin was confirmed by western immunoblotting. Using a solid-phase binding assay, the mGluR1 α-tubulin interaction was shown to be direct, specific, and saturable with a KD of 2.3 ± 0.4 µM. In addition, mGluR1 α, but not mGluR2/3 or mGluR4, could be coimmunoprecipitated from solubilized brain extracts with tubulin using anti-ß-tubulin antibodies. However, mGluR1 α could not be coimmunoprecipitated with the tubulin binding protein gephyrin, nor could it be coimmunoprecipitated with PSD95. Collectively these data demonstrate that the last 86 amino acids of the carboxyl-terminal tail of mGluR1 α are sufficient to determine its interaction with tubulin and that there is an association of this receptor with tubulin in rat brain.

Antimicrobial susceptibility of community-acquired lower respiratory tract bacterial pathogens isolated in the UK during the 1995-1996 cold season.

The antimicrobial susceptibility of 1078 isolates of Haemophilus influenzae, 348 Streptococcus pneumoniae and 258 Moraxella catarrhalis was determined. Overall 15.1% of H. influenzae produced beta-lactamase; 98.8% were susceptible to co-amoxiclav, 85.8% to cefaclor, 96% to clarithromycin and 100% to ciprofloxacin. The majority (94.2%) of M. catarrhalis produced beta-lactamase. The overall prevalence of low-level penicillin resistance (MIC = 0.12-1 mg/L) amongst isolates of S. pneumoniae was 3.4% and that of high-level resistance (MIC > or = 2 mg/L) was 3.7%. Most (96.3%) of the isolates of S. pneumoniae were susceptible to amoxycillin (MIC < or = 0.5 mg/L), 96% to cefaclor (MIC < or = 8 mg/L), 90.7% to clarithromycin (MIC < or = 0.25 mg/L) and 89% to ciprofloxacin (MIC < or = 1 mg/L).