The role of the membrane electrochemical potential in the translocation of acidic and basic residues across the membrane was investigated with the M13 procoat protein, which has a short periplasmic loop, and leader peptidase, which has an extended periplasmically located N‐terminal tail. For both proteins we find that the membrane potential promotes membrane transfer only when negatively charged residues are present within the translocated domain. When these residues are substituted by uncharged amino acids, the proteins insert into the membrane independently of the potential. In contrast, when a positively charged residue is present within the N‐terminal tail of leader peptidase, the potential impedes translocation of the tail domain. However, an impediment was not observed in the case of the procoat protein, where positively charged residues in the central loop are translocated even in the presence of the membrane potential. Intriguingly, several of the negatively charged procoat proteins required the SecA and SecY proteins for optimal translocation. The studies reported here provide insights into the role of the potential in membrane protein assembly and suggest that electrophoresis can play an important role in controlling membrane topology.

No abstract

Previously we have shown that the first hydrophobic domain of leader peptidase (lep) can function to translocate a short N‐terminal 18 residue antigenic peptide from the phage Pf3 coat protein across the plasma membrane of Escherichia coli. We have now examined the mechanism of insertion of N‐terminal periplasmic tails and have defined the features needed to translocate these regions. We find that short tails of up to 38 residues are efficiently translocated in a SecA‐ and SecY‐independent manner while longer tails are very poorly inserted. Efficient translocation of a 138 residue tail is restored and is Sec‐dependent by the addition of a leader sequence to the N‐terminus of the protein. We also find that while there is no amphiphilic helix requirement for N‐terminal translocation, there is a charge requirement that is needed within the tail; an arginine and lysine residue can inhibit or completely block translocation when introduced into the tail region. Intriguingly, the membrane potential is required for insertion of a 38 residue tail but not for a 23 residue tail.

No abstract

Type I signal peptidase (SPase I) catalyzes the cleavage of the amino-terminal signal sequences from preproteins destined for cell export. Preproteins contain a signal sequence with a positively charged n-region, a hydrophobic h-region, and a neutral but polar c-region. Despite having no distinct consensus sequence other than a commonly found c-region "Ala-X-Ala" motif preceding the cleavage site, signal sequences are recognized by SPase I with high fidelity. Remarkably, other potential Ala-X-Ala sites are not cleaved within the preprotein. One hypothesis is that the source of this fidelity is due to the anchoring of both the SPase I enzyme (by way of its transmembrane segment) and the preprotein substrate (by the h-region in the signal sequence) in the membrane. This limits the enzyme-substrate interactions such that cleavage occurs at only one site. In this work we have, for the first time, successfully isolated Bacillus subtilis type I signal peptidase (SipS) and a truncated version lacking the transmembrane domain (SipS-P2). With purified full-length as well as truncated constructs of both B. subtilis and Escherichia coli (Lep) SPase I, in vitro specificity studies indicate that the transmembrane domains of either enzyme are not important determinants of in vitro cleavage fidelity, since enzyme constructs lacking them reveal no alternate site processing of pro-OmpA nuclease A substrate. In addition, experiments with mutant pro-OmpA nuclease A substrate constructs indicate that the h-region of the signal peptide is also not critical for substrate specificity. In contrast, certain mutants in the c-region of the signal peptide result in alternate site cleavage by both Lep and SipS enzymes.

Indoor airborne culturable fungi exposure has been closely linked to occupants' health. However, conventional measurement of indoor airborne fungal concentration is complicated and usually requires around one week for fungi incubation in laboratory. To provide an ultra-fast solution, here, for the first time, a knowledge-based machine learning model is developed with the inputs of indoor air quality data for estimating the concentration of indoor airborne culturable fungi. To construct a database for statistical analysis and model training, 249 data groups of air quality indicators (concentration of indoor airborne culturable fungi, indoor/outdoor PM and PM concentrations, indoor temperature, indoor relative humidity, and indoor CO concentration) were measured from 85 residential buildings of Baoding (China) during the period of 2016.11.15-2017.03.15. Our results show that artificial neural network (ANN) with one hidden layer has good prediction performances, compared to a support vector machine (SVM). With the tolerance of ± 30%, the prediction accuracy of the ANN model with ten hidden nodes can at highest reach 83.33% in the testing set. Most importantly, we here provide a quick method for estimating the concentration of indoor airborne fungi that can be applied to real-time evaluation.

It is of great significance to achieve the prediction of building energy consumption. However, machine learning, as a promising technique for many practical applications, was rarely utilized in this field. The most important reason is that the predictive structure with best performance is difficult to be determined. To fill the gap, this paper offers one in-depth review, which focuses on the accuracy analyses and model comparisons. Specifically, the accuracy analyses were conducted based on different types of buildings (e.g. residential building, commercial building, government building or educational building), different type of temporal granularity (e.g. sub-hourly, hourly, daily or annual), as well as input/output variables and historical data collections. Further, artificial neural network (ANN) and support vector machine (SVM), as the epidemic models, were compared in terms of their complexity of prediction processes, accuracies of results, the amounts of required historical data, the numbers of inputs, etc. Then the hybrid and single machine learning methods were outlined and compared in terms of their strengths and weaknesses. In addition, several vital defects and further research directions are presented from a multivariate perspective. We hope

No abstract