Exploring the molecular linkage of protein stability traits for enzyme optimization by iterative truncation and evolution

The stability of proteins is paramount for their therapeutic and industrial use and, thus, is a major task for protein engineering. Several types of chemical and physical stabilities are desired, and discussion revolves around whether each stability trait needs to be addressed separately and how specific and compatible stabilizing mutations act. We demonstrate a stepwise perturbation-compensation strategy, which identifies mutations rescuing the activity of a truncated TEM β-lactamase. Analyses relating structural stress with the external stresses of heat, denaturants, and proteases reveal our second-site suppressors as general stability centers that also improve the full-length enzyme. A library of lactamase variants truncated by 15 N-terminal and three C-terminal residues (Bla-NΔ15CΔ3) was subjected to activity selection and DNA shuffling. The resulting clone with the best in vivo performance harbored eight mutations, surpassed the full-length wild-type protein by 5.3 °C in T(m), displayed significantly higher catalytic activity at elevated temperatures, and showed delayed guanidine-induced denaturation. The crystal structure of this mutant was determined and provided insights into its stability determinants. Stepwise reconstitution of the N- and C-termini increased its thermal, denaturant, and proteolytic resistance successively, leading to a full-length enzyme with a T(m) increased by 15.3 °C and a half-denaturation concentration shifted from 0.53 to 1.75 M guanidinium relative to that of the wild type. These improvements demonstrate that iterative truncation-optimization cycles can exploit stability-trait linkages in proteins and are exceptionally suited for the creation of progressively stabilized variants and/or downsized proteins without the need for detailed structural or mechanistic information.

In vitro cell quality of articular chondrocytes assigned for autologous implantation in dependence of specific patient characteristics

OBJECTIVE

Autologous chondrocyte implantation (ACI) is a well-established therapeutic option for the treatment of cartilage defects of the knee joint. Since information concerning the cellular aspects of ACI is still limited, the aim of the present study was to investigate relevant differences between chondrocyte quality after in vitro cultivation and possible correlations with patient-specific factors.

DESIGN

Cell quality of 252 consecutive ACI patients was assessed after chondrocyte in vitro expansion by determination of the expression of cartilage relevant surface marker CD44 and cartilage-specific differentiation markers (aggrecan and collagen type II). All cell quality parameters were correlated with patient-specific parameters, such as age, size and defect location, number of defects and grade of joint degeneration according to the Kellgren-Lawrence classification.

RESULTS

Neither the expression of CD44, aggrecan or collagen type II nor cell density or viability after proliferation seemed to correlate with the grade of joint degeneration, defect aetiology or patient gender. However, chondrocytes harvested from the knee joints of patients at less than 20 years of age showed significantly higher expression rates of cartilage-specific markers when compared to older patients' chondrocytes.

CONCLUSIONS

The present study identifies relevant differences concerning chondrocyte quality after in vitro expansion in a highly preselected study population of 252 patients that from a surgical point of view were eligible for ACI. With the exception of patients aged 20 years or younger, no patient-specific parameters could be identified which might allow anticipation of cell quality in individual patients.

A general method of terminal truncation, evolution, and re-elongation to generate enzymes of enhanced stability

Improving enzyme stability is a highly desirable design step in generating enzymes able to function under extreme conditions, such as elevated temperatures, while having the additional benefit of being less susceptible to cleavage by proteases. For these reasons, many different approaches and techniques have been devised in constructing such proteins, but the results to date have been of mixed success. Here, we present a robust method involving the terminal truncation, random mutagenesis and fragmentation, recombination, elongation, and finally, selection at physiological temperatures, to generate an enzyme with improved stability. Three cycles of directed evolution comprising of random mutagenesis, DNA shuffling, and selection at 37 degrees C were used, using the bacterial enzyme TEM-1 beta-lactamase as a model protein to yield deletion mutants with in vivo ampicillin resistance levels comparable to wild-type (wt) enzyme. Kinetic studies demonstrate the selected mutant to have a significantly improved thermostability relative to its wt counterpart. Elongation of this mutant to the full-length gene resulted in a beta-lactamase variant with dramatically increased thermostability. This technique was so fruitful that the evolved enzyme retained its maximum catalytic activity even 20 degrees C above its wt parent protein optimum. Thus, structural perturbation by terminal truncation and subsequent compensation by directed evolution at physiological temperatures is a fast, efficient, and highly effective way to improve the thermostability of proteins without the need for selecting at elevated temperatures.

Structural perturbation and compensation by directed evolution at physiological temperature leads to thermostabilization of beta-lactamase

The choice of protein for use in technical and medical applications is limited by stability issues, making understanding and engineering of stability key. Here, enzyme destabilization by truncation was combined with directed evolution to create stable variants of TEM-1 beta-lactamase. This enzyme was chosen because of its implication in prodrug activation therapy, pathogen resistance to lactam antibiotics, and reporter enzyme bioassays. Removal of five N-terminal residues generated a mutant which did not confer antibiotic resistance at 37 degrees C. Accordingly, the half-life time in vitro was only 7 s at 40 degrees C. However, three cycles comprising random mutagenesis, DNA shuffling, and metabolic selection at 37 degrees C yielded mutants providing resistance levels significantly higher than that of the wild type. These mutants demonstrated increased thermoactivity and thermostability in time-resolved kinetics at various temperatures. Chemical denaturation revealed improved thermodynamic stabilities of a three-state unfolding pathway exceeding wild-type construct stability. Elongation of one optimized deletion mutant to full length increased its stability even further. Compared to that of the wild type, the temperature optimum was shifted from 35 to 50 degrees C, and the beginning of heat inactivation increased by 20 degrees C while full activity at low temperatures was maintained. We attribute these effects mainly to two independently acting boundary interface residue exchanges (M182T and A224V). Structural perturbation by terminal truncation, evolutionary compensation at physiological temperatures, and elongation is an efficient way to analyze and improve thermostability without the need for high-temperature selection, structural information, or homologous proteins.