Software-assisted automated detection and identification of "unknown" fentanyl analogues.

Fentanyl and its non-pharmaceutical analogues (NPFs) are potent synthetic opioids, traditionally used for pain management, with ever-increasing illicit uses. Tightening the regulation for known fentanyls leads to new synthetic analogues in the opioid market. Furthermore, the Organization for the Prohibition of Chemical Weapons (OPCW) has recently issued a decision regarding aerosolized use of central nervous system (CNS)-acting agents, such as fentanyl and its analogues, under the concern that these materials could be misused for terror or war purposes. The ever-increasing development of new fentanyl analogues makes the task of detection and identification of these new, unknown analogues crucial. In this work, we introduce an automated tool for the detection and putative identification of "unknown" fentanyl analogues, using liquid chromatography-mass spectrometry (LC-MS) (high-resolution mass spectrometry [HRMS]) analysis, subsequently followed by data processing using the "Compound Discoverer" software. This software, in our modified use, enabled the automatic detection of various fentanyl analogues, by "digging" out components and comparing them to pre-calculated theoretical molecular ions of possible modifications or transformations on the fentanyl backbone structure (no library or database used). Subsequently, structural elucidation for the proposed component of interest is carried out by automated MS/MS data interpretation, as performed by the software. This method was explored on 12 fentanyl-based "unknown" analogues used as model examples, including chemical modifications such as fluorination and methylation. In all tested compounds, automatic detection and identification were achieved, even at concentrations as low as 1 ng/mL in an environmental soil matrix extract.

Monitoring Exposure to Five Chemical Warfare Agents Using the Dried Urine Spot Technique and Liquid Chromatography-Mass Spectrometry/Mass Spectrometry-In Vivo Determination of Sarin Metabolite in Mice.

Dried urine spot (DUS) is a micro-sample collection technique, known for its advantages in handling, storage and shipping. It also uses only a small volume of urine, an essential consideration in working with small animals, or in acute medical situations. Alkyl-phosphonic acids are the direct and indicative metabolites of organophosphorus chemical warfare agents (OP-CWAs) and are present in blood and urine shortly after exposure. They are therefore crucially important for monitoring casualties in war and terror scenarios. We report here a new approach for the determination of the metabolites of five CWAs in urine using DUS. The method is based on a simple and rapid sample preparation, using only 50 µL of urine, spotted and dried on DBS paper, extracted using 300 µL methanol/water and analyzed via targeted LC-MS/MS. The detection limits for the five CWAs, sarin (GB), soman (GD), cyclosarin (GF), VX and RVX in human urine were from 0.5 to 5 ng/mL. Recoveries of (40-80%) were obtained in the range of 10-300 ng/mL, with a linear response (R2 > 0.964, R > 0.982). The method is highly stable, even with DUS samples stored up to 5 months at room temperature before analysis. It was implemented in a sarin in vivo exposure experiment on mice, applied for the time course determination of isopropyl methylphosphonic acid (IMPA, sarin hydrolysis product) in mice urine. IMPA was detectable even with samples drawn 60 h after the mice's (IN) exposure to 1 LD50 sarin. This method was also evaluated in a non-targeted screening for multiple potential CWA analogs (LC-Orbitrap HRMS analysis followed by automatic peak detection and library searches). The method developed here is applicable for rapid CWA casualty monitoring.

Evaluation of Matrix Complexity in Nontargeted Analysis of Small-Molecule Toxicants by Liquid Chromatography-High-Resolution Mass Spectrometry.

Complex mixtures, characterized by high density of compounds, challenge trace detection and identification. This is further exacerbated in nontargeted analysis, where a compound of interest may be well hidden under thousands of matrix compounds. We studied the effect of matrix complexity on nontargeted detection (peak picking) by LC-MS/MS (Orbitrap) analysis. A series of ∼20 drugs, V-type chemical warfare agents and pesticides, simulating toxic unknowns, were spiked at various concentrations in several complex matrices including urine, rosemary leaves, and soil extracts. Orbitrap "TraceFinder" software was used to explore their peak intensities in relation to the matrix (peak location in an intensity-sorted list). Average practical detection limits of nontargets were determined. While detection among the first 10,000 peaks was achieved at 0.3-1 ng/mL levels in the extract, for the more realistic "top 1000" list, much higher concentrations were required, approaching 10-30 ng/mL. A negative power law functional dependence between the peak location in an intensity-sorted suspect list and the nontarget concentration is proposed. Controlled complexity was explored with a series of urine dilutions, resulting in an excellent correlation between the power law coefficient and dilution factor. The intensity distribution of matrix peaks was found to spread (unevenly) on a broad range, fitting well the Weibull distribution function with all matrices and extracts. The quantitative approach demonstrated here gives a measure of the actual capabilities and limitations of LC-MS in the analysis of nontargets in complex matrices. It may be used to estimate and compare the complexity of matrices and predict the typical detection limits of unknowns.

Integrating Effect-Directed Analysis and Chemically Indicative Mass Spectral Fragmentation to Screen for Toxic Organophosphorus Compounds.

Analytical chemists are often challenged to screen for bioactive compounds in complex matrices, sometimes without a priori knowledge of the exact compound of interest. Therefore, "flagging" techniques, highlighting common characteristics of bioactive compounds, are highly sought after. In this work, we demonstrate a double flagging method, where unknown organophosphorus acetylcholinesterase inhibitors are "flagged" out of a complex matrix by the presence of organophosphorus-indicative ions as well as their acetylcholinesterase inhibition. This is accomplished by flagging the LC chromatographic retention time of phosphorus-indicative ions using accurate mass high-energy in-source CID products, and the retention time of acetylcholinesterase inhibiting compounds using a parallel microfractionation-based bioassay. We successfully apply this method to screen VX, VM, and RVX nerve agents as well as methomyl, a carbamate pesticide, out of soil and whole blood samples at low µM to sub-µM concentrations. This methodology can be easily extended to diverse chemical families and biological activities of interest.

Software-Assisted Automated Detection and Identification of "Unknown" Analogues: Implementation on V-Type Nerve Agents.

V-type nerve agents are among the most toxic organophosphorus chemical warfare agents, and they are under strict regulation and supervision by the OPCW (Organization for the Prohibition of Chemical Weapons). The V-type class of materials refers to a potentially large number of analogues and isomers. In order to expose instances of unfulfillment of the OPCW treaty, it is essential to have the ability to detect and identify "unknown" analogues of this family, even in the absence of an analytical standard. This work demonstrates a new automated tool for the detection and identification of V-type analogues, using high-resolution-accurate-mass LC-MS analysis, followed by "Compound Discoverer" software data processing. This software, originally developed for metabolism and metabolomics screening, is used here to automatically detect various V-type analogues by picking peaks and comparing them to "in-silico" calculated modifications made on a predefined basic V-backbone structure (according to the OPCW definitions for V-type agents). Subsequently, a complete structural elucidation for the proposed molecular formula is obtained by MS/MS data analysis of the suspected component, for both the V-type analogue (using ESI(+) analysis) as well as its hydrolysis product (using ESI(-) analysis) for a better elucidation of the phosphonate "head" structure. This method was found to be useful for the detection and identification of several "unknown" analogues, at low ng/mL levels in soil extracts.

pyAIR-A New Software Tool for Breathomics Applications-Searching for Markers in TD-GC-HRMS Analysis.

Volatile metabolites in exhaled air have promising potential as diagnostic biomarkers. However, the combination of low mass, similar chemical composition, and low concentrations introduces the challenge of sorting the data to identify markers of value. In this paper, we report the development of pyAIR, a software tool for searching for volatile organic compounds (VOCs) markers in multi-group datasets, tailored for Thermal-Desorption Gas-Chromatography High Resolution Mass-Spectrometry (TD-GC-HRMS) output. pyAIR aligns the compounds between samples by spectral similarity coupled with retention times (RT), and statistically compares the groups for compounds that differ by intensity. This workflow was successfully tested and evaluated on gaseous samples spiked with 27 model VOCs at six concentrations, divided into three groups, down to 0.3 nL/L. All analytes were correctly detected and aligned. More than 80% were found to be significant markers with a p-value < 0.05; several were classified as possibly significant markers (p-value < 0.1), while a few were removed due to background level. In all group comparisons, low rates of false markers were found. These results showed the potential of pyAIR in the field of trace-level breathomics, with the capability to differentially examine several groups, such as stages of illness.

A multiple-method comparative study using GC-MS, AMDIS and in-house-built software for the detection and identification of "unknown" volatile organic compounds in breath.

The human respiratory system is a highly complex matrix that exhales many volatile organic compounds (VOCs). Breath-exhaled VOCs are often "unknowns" and possess low concentrations, which make their analysis, peak digging and data processing challenging. We report a new methodology, applied in a proof-of-concept experiment, for the detection of VOCs in breath. For this purpose, we developed and compared four complementary analysis methods based on solid-phase microextraction and thermal desorption (TD) tubes with two GC-mass spectrometer (MS) methods. Using eight model compounds, we obtained an LOD range of 0.02-20 ng/ml. We found that in breath analysis, sampling the exhausted air from Tedlar bags is better when TD tubes are used, not only because of the preconcentration but also due to the stability of analytes in the TD tubes. Data processing (peak picking) was based on two data retrieval approaches with an in-house script written for comparison and differentiation between two populations: sick and healthy. We found it best to use "raw" AMDIS deconvolution data (.ELU) rather than its NIST (.FIN) identification data for comparison between samples. A successful demonstration of this method was conducted in a pilot study (n = 21) that took place in a closed hospital ward (Covid-19 ward) with the discovery of four potential markers. These preliminary findings, at the molecular level, demonstrate the capabilities of our method and can be applied in larger and more comprehensive experiments in the omics world.

Extended retrospective detection of regenerated sarin (GB) in rabbit blood and the IMPA metabolite in urine: a pharmacokinetics study.

Long-term retrospective monitoring of exposure to organophosphorus nerve agents is challenging. We recently developed two highly sensitive analytical methods for regenerated sarin (GB) nerve agent in blood and its primary metabolite, isopropyl-methylphosphonic acid (IMPA), in urine. These methods were implemented in a toxicokinetics study carried out with sarin injected (i.v.) to rabbits at doses corresponding to 0.1, 0.5 or 0.9 LD50. The time frame for monitoring regenerated sarin from blood was 70 days for 0.1 LD50 and 0.5 LD50 and 77 days for 0.9 LD50, where rapid elimination occurred in the first 8 days with an initial average half-life of 1.2 days, followed by a second, slower elimination, with a terminal average half-life of 8.4 days. The time frame for monitoring IMPA in urine was 7, 15 and 16 days for 0.1 LD50, 0.5 LD50 and 0.9 LD50 intoxications, respectively. Rapid elimination of IMPA in urine occurred after exposure, with an average half-life of ~ 0.8 days on days 2-6. For the first time, a slower elimination route for IMPA, with an average half-life of ~ 4 days from day 6 onwards, was revealed. Both IMPA and regenerated sarin pharmacokinetics exhibit linearity with dose. The overlaid pharmacokinetic profiles of regenerated sarin in blood along with IMPA in urine emphasize the dominance of IMPA with a rapid decay in urine in the first week and the slower long-term decay of protein-bound sarin later in blood. To our knowledge, the two new sensitive methods exhibit the longest monitoring time frame reported in biological samples.

Unraveling mosquito metabolism with mass spectrometry-based metabolomics.

Nearly half a million people die annually due to mosquito-borne diseases. Despite aggressive mosquito population-control efforts, current strategies are limited in their ability to control these vectors. A better understanding of mosquito metabolism through modern approaches can contribute to the discovery of novel metabolic targets and/or regulators and lead to the development of better mosquito-control strategies. Currently, cutting-edge technologies such as gas or liquid chromatography-mass spectrometry-based metabolomics are considered 'mature technologies' in many life-science disciplines but are still an emerging area of research in medical entomology. This review primarily discusses recent developments and progress in the application of mass spectrometry-based metabolomics to answer multiple biological questions related to mosquito metabolism.

Enantioselective in-vitro elimination kinetics of nerve agents in blood monitored by derivatization and LC-MS/MS analysis.

We present a simple method for chiral separation and analysis of organophosphorus nerve agents and apply it to monitor the enantioselective blood elimination kinetics of sarin in-vitro. The method is implemented in standard reverse phase LC-MS operating conditions, relieving the user of the dedicated operating conditions frequently demanded in chiral LC-MS analysis. The method consists of formation of diastereomers by a rapid derivatization with (R)-2-(1 aminoethyl) phenol, followed by LC-MS/MS analysis. Derivatization enantioselectivity was studied by comparing the reaction of optically pure sarin and racemic sarin, proving no substantial enantiomeric preference in the reaction and demonstrating the enantiomeric discrimination abilities of the technique. Enantioselective sarin elimination pathways were probed in-vitro by following the fast elimination kinetics of the two sarin enantiomers as well as its hydrolysis metabolite (isopropyl methyl-phosphonic acid, IMPA) in whole blood and plasma compared to water. Sarin enantiomers showed the known marked differences in elimination kinetics with rapid elimination of the (+) enantiomer and slower elimination of the (-) enantiomer in whole blood and plasma as well as dose-dependent kinetics (faster elimination at lower concentrations). We found that small amounts of acetonitrile in plasma prevent the rapid elimination of the (+) enantiomer, resulting in similar, slower elimination kinetics for both enantiomers.

Highly sensitive retrospective determination of organophosphorous nerve agent biomarkers in human urine implemented in vivo in rabbit.

Highly toxic organophosphorous nerve agents (OPAs) have been used in several armed conflicts and terror attacks in the last few decades. A new method for retrospective determination of alkyl methylphosphonic acid (AMPA) metabolites in urine after exposure to VX, GB and GF nerve agents was developed. This method enables a rapid, sensitive and selective determination of trace levels of the nerve agent biomarkers ethyl methylphosphonic acid (EMPA), isopropyl methylphosphonic acid (IMPA) and cyclohexyl methylphosphonic acid (CMPA) in urine. The new technique involves a unique combination of two solid phase extraction (SPE) cartridges: a Ba/Ag/H cartridge for urine interference removal, and a ZrO2 cartridge for selective reconstitution and enrichment of the AMPAs. Extraction of AMPAs from the ZrO2 cartridge was accomplished with a 1% ammonium hydroxide (NH4OH) solution and was followed by analysis via liquid chromatography-mass spectrometry (LC-MS). The limits of quantitation (LOQs) were in the range of 10-100 pg/mL with recoveries of 64-71% (± 5-19%) after fast sample preparation and a total LC-MS analysis cycle time of 15 min and 13 min, respectively. This method was successfully applied in vivo in a rabbit that was exposed to 0.5 LD50 (7.5 µg/kg, i.v.) sarin for retrospective monitoring of the IMPA metabolite in urine. For the first time, IMPA was determined in rabbit urine samples for 15 days post-exposure, which is longer than any reported post-exposure method for AMPAs. To the best of our knowledge, this new method is the most sensitive and rapid for AMPA determination in urine by LC-MS/MS analysis.

Mass spectrometry-based stable-isotope tracing uncovers metabolic alterations in pyruvate kinase-deficient Aedes aegypti mosquitoes.

A recent in vitro characterization of a recombinant pyruvate kinase (PK) from Aedes aegypti mosquitoes demonstrated that the enzyme is uniquely regulated by multiple allosteric effectors. Here, we further explored PK gene and protein expression, and enzymatic activity in key metabolic tissues of mosquitoes maintained under different nutritional conditions. We also studied the metabolic effects of PK depletion using several techniques including RNA interference and mass spectrometry-based stable-isotope tracing. Transcriptional analysis showed a dynamic post-feeding PK mRNA expression pattern within and across mosquito tissues, whereas corresponding protein levels remained stable throughout the time course analyzed. Nevertheless, PK activity significantly differed in the fat body of sucrose-, blood-fed, and starved mosquitoes. Genetic silencing of PK did not alter survival in blood-fed females maintained on sucrose. However, an enhanced survivorship was observed in PK-deficient females maintained under different nutritional regimens. Our results indicate that mosquitoes overcame PK deficiency by up-regulating the expression of genes encoding NADP-malic enzyme-1, phosphoenolpyruvate carboxykinase-1, phosphoglycerate dehydrogenase and glutamate dehydrogenase, and by decreasing glucose oxidation and metabolic pathways associated with ammonia detoxification. Taken together, our data demonstrate that PK confers to A. aegypti a metabolic plasticity to tightly regulate both carbon and nitrogen metabolism.

Instantaneous monitoring of free sarin in whole blood by dry blood spot-thermal desorption-GC-FPD/MS analysis.

Dry blood spot (DBS), a micro whole-blood sampling technique, enables rapid and self-blood collection; it is stable and economical. Currently, DBS filters require various sample preparation procedures specifically tailored for the target compounds, which are followed by GC-MS or LC-MS analysis. However, the small amounts of blood make the approach analytically challenging, mostly in terms of sensitivity and quantification. Herein, we introduce a new DBS concept for GC-compatible volatile to semi-volatile compounds in which DBS is directly coupled with thermal desorption analysis, thus eliminating time consuming treatments. Furthermore, to stabilize the target compound over the sampling DBS substrate, a commercial filter based on an extremely efficient trapping adsorption phase, styrene-divinylbenzene (SDVB), is first used. The performance of the new SDVB-DBS concept was demonstrated herein for monitoring the most volatile chemical warfare agent, sarin, which might be present in blood and the detection of which is usually challenging due to its rapid metabolism. This study encompasses adequate sampling and analysis method parametrization and validation, leading to a detection sensitivity of 100 pg sarin per 30 µL whole blood in 5-day-old samples, with a linear dynamic range of two orders of magnitude, adequate precision, and acceptable accuracy. Applying the method to an in-vivo mouse intranasal exposure experiment (3LD50 GB) enabled the successful detection of 25-90 ng mL-1 free sarin in blood samples drawn 2 min after exposure. The method's performance clearly emphasizes the potential of the new concept in "freezing the clock" for reactive whole blood media in pharmacokinetics and pharmacodynamics studies, as well as in applications in which informative and reliable monitoring of unstable target compounds and biomarkers is desired.

Retrospective determination of regenerated nerve agent sarin in human blood by liquid chromatography-mass spectrometry and in vivo implementation in rabbit.

The highly toxic nerve agent sarin (o-isopropyl methyl-phosphonofluoridate, GB) has been used in several armed conflicts and terror attacks in recent decades. Due to its inherent high sensitivity, liquid chromatography-mass spectrometry (LC-MS/MS) has the potential to detect ultratrace levels of fluoride-regenerated G and V agents after appropriate chemical derivatization. A new method for the retrospective determination of exposure to sarin was developed. The method is based on sarin regeneration from blood using the fluoride-induced technique followed by derivatization with 2-[(dimethylamino)methyl]phenol (2-DMAMP) and LC-ESI-MS/MS (MRM) analysis. The validated method presents good linear response in the concentration range of 5-1000 pg/mL with a limit of quantitation (LOQ) of 5 pg/mL, 13.8% accuracy, 16.7% precision and a total recovery of 62% ± 9%. This new analytical approach has several advantages over existing GC/GC-MS-based methods in terms of sensitivity, specificity and simplicity, in addition to a short LC-MS cycle time of 12 min. The method was successfully applied in an in vivo experiment for retrospective determination of sarin in a rabbit exposed to 0.1 LD50 sarin (1.5 µg/kg, i.v.). GB-2-DMAMP was easily determined in samples drawn up to 11 days after exposure. The high S/N ratio (500) observed for the GB-2-DMAMP signal in the 11day sample poses the potential for an extended time frame of months for analysis with this new method for the retrospective detection of sarin exposure. To the best of our knowledge, this is the first report on LC-MS/MS trace analysis of regenerated GB from biological matrices.

Effective neutralization of chemical warfare agents (HD, VX) by Me-DABCOF: a small molecule with dual action.

The ability of mono N-methyl-1,4-diazabicyclo[2.2.2]octane fluoride (Me-DABCOF, 1) to act as a bifunctional reagent that effectively and universally neutralizes both the persistent and extremely toxic blister agent HD and the nerve agent VX in nearly neutral aqueous solution, alumina powder or a hydrogel formulation, is described.

Identification of G-nerve agents at picogram levels from complex organic samples containing hydrocarbon interferences by aqueous extraction, followed by derivatization and liquid chromatography-mass spectrometry analysis.

The chromatograms obtained from the gas chromatography-electron ionization mass spectrometric (GC-EI-MS) analysis of extracts containing G-nerve agents in the presence of diesel, gasoline, etc., are dominated by hydrocarbon backgrounds that "mask" the G-nerve agents, leading to severe difficulties in identification. This paper presents a practical solution for this challenge by transferring the G-nerve agents from the organic phase into the aqueous phase using liquid-liquid extraction (LLE), followed by derivatization with 2-[(dimethylamino)methyl]phenol (2-DMAMP), allowing ultrasensitive LC-ESI-MS/MS analysis of the G-derivatives. The proposed approach enables rapid identification of trace amounts of G-nerve agents with limits of identification (LOIs) at the pg/mL scale.

Aqueous extraction followed by derivatization and liquid chromatography-mass spectrometry analysis: A unique strategy for trace detection and identification of G-nerve agents in environmental matrices.

A highly sensitive method for the detection and identification of sarin (GB), soman (GD) and cyclosarin (GF) chemical warfare agents (CWAs) in environmental outdoor and indoor matrices such as soil, asphalt, linoleum, formica, concrete and cloth was developed. The method incorporates derivatization of the G-type nerve agent extracts with 2-[(dimethylamino)methyl]phenol (2-DMAMP), followed by LC-ESI(+)-MS/MS analysis. Four LC-amenable extraction solvents were explored in terms of their extraction efficiency and the reaction rate of the derivatizing agent. The reaction time, temperature and derivatization reagent amount were optimized. The optimal procedure was found to be extraction with water by agitation (2 min), followed by the addition of 2-DMAMP directly into the injection vial and stirring for 5 min prior to LC-ESI(+)-MS/MS analysis, without any other pretreatment. The method was applied to real-world samples and exhibited very low detection limits (LODs) of 0.8-20 pg/cm2 in asphalt, linoleum, cloth, formica and concrete and 4 pg/g in soil. The newly developed method demonstrated significantly superior sensitivity compared to conventional GC-MS- and LC-MS-based methods for the identification of G-nerve agents and allowed the determination of both G-nerve agents and their hydrolysis products within a single LC-MS/MS run. The proposed methodology may be practical for verifying contaminated matrices collected in the battlefield or terror scenes in forensic investigations where trace level analysis is required.

Dry Blood Spot sample collection for post-exposure monitoring of chemical warfare agents - In vivo determination of phosphonic acids using LC-MS/MS.

Phosphonic acids are the direct and immediate metabolites of organophosphorus chemical warfare agents (OP-CWAs). Accordingly, their detection serves for evaluating exposure to OP-CWAs in a terror or war scenario. After exposure, phosphonic acids are present in the blood; however, blood drawing must be carried out by medical personnel, hence the number of samples that can be drawn in a mass-casualty event is limited. Herein, we describe a new approach developed for the determination of phosphonic acids in blood using Dry Blood Spots (DBSs) on a filter paper. The method is based on a simple sample preparation protocol, followed by LC-MS-MS targeted (MRM) analysis. The detection limits of Soman (GD), Cyclosarin (GF) and VX metabolites in whole blood were as low as 1 ng/ml, while the detection limits were 0.3 ng/ml for the GF metabolite and 0.5 ng/ml for the Sarin (GB) metabolite. Good recoveries were obtained in the range of 1-100 ng/ml for GB and GD metabolites, and 3-100 ng/ml for GF, VX and RVX metabolites, with a linear response (R2 = 0.99). The method has proven to be reliable even with DBS samples stored up to 35 days at room temperature before analysis. This method was implemented in a 24 h time-course determination of the Sarin metabolite in an in - vivo experiment, after rat exposure to 1 LD50 of Sarin. This technique is simple, rapid, sensitive, robust, long lasting and compatible with field collection and storage; hence, it can serve for large-scale sampling and reliable monitoring of potential OP-CWAs casualties. Since DBS sampling is amenable to nonprofessionals, including self-sampling, this technique is highly suitable for mass-casualty incidents.

Oxidation-assisted structural elucidation of compounds containing a tertiary amine side chain using liquid chromatography mass spectrometry.

A novel analytical technique for the structural elucidation of compounds bearing a tertiary amine side chain via "in vial" instantaneous oxidation and liquid chromatography mass spectrometry (LC-MS) was developed. A series of lidocaine homologs and benzimidazole derivatives with a major/single amine representative base peak in both their EI-MS and ESI-MS/MS spectra were subjected to oxidation by a 0.1% solution of hydrogen peroxide (including several 16 O/18 O exchange experiments), followed by LC-ESI-MS/MS analysis. The N-oxide counterparts promoted extensive fragmentation with complete coverage of all parts of the molecule, enabling detailed structural elucidation and unambiguous identification of the unoxidized analytes at low nanogram per milliliter levels.