Human proteome

Two separate research groups have developed the first draft maps of the human proteome, a landmark akin to the first mapping of the human genome in 2003. Both studies were published in the same issue of the journal Nature (Nature 2014;509:575–81; Nature 2014;509:582–87).

An international team from the United States, India, Chile, London, and Hong Kong used liquid chromatography tandem mass spectrometry to systematically identify and annotate proteins encoded by 17,294 genes, accounting for about 84% of the total annotated protein-coding genes in humans. This team conducted in-depth proteomic analysis from 30 histologically normal human tissue samples, including 17 from adults, seven fetal tissues, and six purified hematopoietic cells. The researchers have published their findings in a catalogue, available online at

This team confirmed 223,385 exons, 4,105 N termini, and 66,947 exon-exon junctions. They also identified 193 novel protein coding regions.

The second team, from Germany, developed a draft human proteome map by using data already available from repositories or contributed by colleagues, as well as their own analyses. This resulted in annotation of proteins associated with 18,097 of the 19,629 human genes annotated in Swiss-Prot, a curated protein database, or about 92%. For the proteins they themselves analyzed—about 40% of the annotated proteins—the team used ultra-high-pressure liquid chromatography and tandem mass spectrometry to characterize samples from 60 human tissues, 13 body fluids, and 147 cancer cell lines. This team also created ProteomicsDB, a database designed for real-time analysis of big data.

This team also compared messenger RNA and protein expression profiles for 12 human tissues. They found that both mRNA and protein levels "vary greatly" between tissues, but that the ratio between mRNA and protein levels is "remarkably conserved" between tissues for any given protein.

Newborn Screening for Congenital Adrenal Hyperplasia Highly Effective in Detecting Severe Forms of the Disease

A study in Sweden found that the country's 26-year newborn screening program for congenital adrenal hyperplasia (CAH) was highly effective in detecting the salt-wasting form of the disease, thereby reducing mortality (JAMA Pediatr 2014;168:567–75). The authors conducted the study in response to recent reports that had questioned the rationale for CAH newborn screening, because of high recall rates for positive screening results and low-sensitivity­ in detecting salt-wasting forms of CAH.

The authors looked at 2,737,932 CAH newborn screening results between 1986 and 2011, which accounted for 99.8% of all babies born in Sweden during that period. In all, 274 of these infants had CAH. 231 newborns had positive screening results, yielding an overall sensitivity of 84.3% for all forms of CAH, along with 99.9% specificity.

Over time 17-hydroxyprogesterone assay cutoff levels for full-term babies were lowered, raised, then lowered again, and now rest at 60 nmol/L. Likewise, cutoffs for preterm babies varied over time, and currently are 350 nmol/L for those born before 35 weeks' gestation, and 100 nmol/L for those born at 35 or 36 weeks.

All babies were recalled after a positive screening test, at which time a second filter paper sample was obtained. Babies with "convincingly decreased" 17-hydroxyprogesterone levels from the recall test were considered false-positives if they also had no clinical symptoms of CAH.

CYP21A2 mutations cause more than 90% of CAH cases, and in Sweden, 94.8% of babies with true-positive screening results underwent genotyping. Screening levels of 17-hydroxyprogesterone correlated with CYP21A2 mutations, with higher concentrations associated with more severe disease.

In all, 43 cases were missed during the initial newborn screening. Preterm babies missed in the initial screening all had the I172N genotype group, while full-term infants with false-negative­ results belonged primarily to the nonclassic V281L genotype group.

Oral Fluid Cannabinoid Collection Devices Yield Significantly Different Results

A controlled marijuana intake study comparing the performance of two oral fluid collection devices and an onsite screening device found the devices reported significantly different 11-nor-9-Carboxy-THC (THCCOOH) and cannabinol but not Δ 9-tetrahydrocannabinol (THC) concentrations (Anal Bioanal Chem 2014;406:4117–28). The disparate results could affect oral fluid data interpretation, according to the authors.

The authors conducted the study because, cannabis, already the most ommonly used illicit drug worldwide, is being used more in both recreational and medicinal settings, so accurate onsite testing for it is needed. Oral fluid devices offer the advantages of noninvasive, observed sample collection with lower risk of adulteration than urine samples. Oral fluid devices offer rapid turnaround times, objective result interpretation, as well as quality control information. Yet, different device formulations such as buffer dilution, cannabinoid antibodies, and detection mechanisms make inter-method comparisons challenging.

The authors sought to evaluate the performance characteristics and oral fluid detection windows for the on-site DrugTest 5000 and the StatSure and OralEze devices. They compared methods with the participation of 24 frequent or occasional cannabis smokers who, for this study, smoked one 6.8% THC cigarette in 10 minutes. Participants entered a secure research unit approximately 19 hours before smoking. The researchers collected samples upon admission, 1 hour before, and at 15 time points after smoking, ranging from 
30 minutes to 30 hours.

The researchers analyzed test cassettes within 15 minutes, followed by two-dimensional gas chromatography analysis within 24 hours. They found significantly higher concentrations of THCCOOH and cannabinoid in OralEze versus Stat-Sure samples. The DrugTest 5000 had high diagnostic sensitivity, specificity, and efficiency for cannabinoids.

Blood Biomarkers Show Promise in Detecting Sports-Related Brain Injury

A prospective cohort study among professional hockey players who experienced concussion found increased blood levels of total tau (T-tau) and S-100 calcium-binding­ protein B (S-100B) after injury versus at baseline (JAMA Neurol 2014;71:684–92). The findings demonstrate these analytes to be promising for use both in diagnosing concussion and in decision-making for determining whether an athlete can be declared fit to return-to-play, according to the authors.

The researchers examined blood markers of concussion or mild traumatic brain injury because they offer a practical way to evaluate and follow athletes after injury. T-tau, a marker of axonal damage, already has been shown when measured in cerebrospinal fluid (CSF) to be elevated after acute brain injury, with levels correlated to the severity of injury. Similarly, S-100B, a glial cell biomarker, and neuron-specific enolase (NSE), a marker of neuronal injury, have been found at increased CSF levels in individuals with sports injuries.

Based on these prior findings, the authors examined levels of T-tau and serum S-100B and NSE in 288 professional hockey players. All players had baseline clinical examination prior to the start of the 2012–2013 season, and 47 underwent pre-season blood sampling. In all, 35 players experienced a concussion, and of these, 28 had repeated blood sampling at 1, 12, 36, and 144 hours after injury, and at the time they returned to play.

The researchers found that concussed players had increased T-tau and S-100B, compared with baseline. The median post-injury T-tau level was 10 pg/mL compared with 4.5 pg/mL preseason, a statistically significant difference. Pre-season S-100B levels were 0.045 µg/L, compared with 0.075 µg/L post-season, also a statistically significant difference. Levels of both T-tau and S-100B were highest immediately after concussion, and they decreased during rehabilitation. In contrast, the researchers did not find significant pre-season and post-concussion differences in NSE concentrations.