Transmembrane Protease Serine 2 Proteolytic Cleavage of the SARS-CoV-2 Spike Protein: A Mechanistic Quantum Mechanics/Molecular Mechanics Study to Inspire the Design of New Drugs To Fight the COVID-19 Pandemic.

Despite the development of vaccines against the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus, there is an urgent need for efficient drugs to treat infected patients. An attractive drug target is the human transmembrane protease serine 2 (TMPRSS2) because of its vital role in the viral infection mechanism of SARS-CoV-2 by activation of the virus spike protein (S protein). Having in mind that the information derived from quantum mechanics/molecular mechanics (QM/MM) studies could be an important tool in the design of transition-state (TS) analogue inhibitors, we resorted to adiabatic QM/MM calculations to determine the mechanism of the first step (acylation) of proteolytic cleavage of the S protein with atomistic details. Acylation occurred in two stages: (i) proton transfer from Ser441 to His296 concerted with the nucleophilic attack of Ser441 to the substrate's P1-Arg and (ii) proton transfer from His296 to the P1'-Ser residue concerted with the cleavage of the ArgP1-SerP1' peptide bond, with a Gibbs activation energy of 17.1 and 15.8 kcal mol-1, relative to the reactant. An oxyanion hole composed of two hydrogen bonds stabilized the rate-limiting TS by 8 kcal mol-1. An analysis of the TMPRSS2 interactions with the high-energy, short-lived tetrahedral intermediate highlighted the limitations of current clinical inhibitors and pointed out specific ways to develop higher-affinity TS analogue inhibitors. The results support the development of more efficient drugs against SARS-CoV-2 using a human target, free from resistance development.

Pharmacological re-assessment of traditional medicinal plants-derived inhibitors as antidotes against snakebite envenoming: A critical review.

ETHNOPHARMACOLOGICAL RELEVANCE: Traditional healers have used medicinal plants to treat snakebite envenomation worldwide; however, mostly without scientific validation. There have been many studies on the therapeutic potential of the natural products against snake envenomation. AIM OF THE STUDY: This review has highlighted snake venom inhibitory activity of bioactive compounds and peptides from plants that have found a traditional use in treating snakebite envenomation. We have systematically reviewed the scenario of different phases of natural snake venom inhibitors characterization covering a period from 1994 until the present and critically analysed the lacuna of the studies if any, and further scope for their translation from bench to bedside. MATERIALS AND METHODS: The medicinal plant-derived compounds used against snakebite therapy were reviewed from the available literature in public databases (Scopus, MEDLINE) from 1994 till 2020. The search words used were 'natural inhibitors against snakebite,' 'natural products as therapeutics against snakebite,' 'natural products as antidote against snake envenomation,' ' snake venom toxin natural inhibitors,' 'snake venom herbal inhibitors'. However, the scope of this review does not include computational (in silico) predictions without any wet laboratory validation and snake venom inhibitory activity of the crude plant extracts. In addition, we have also predicted the ADMET properties of the identified snake venom inhibitors to highlight their valuable pharmacokinetics for future clinical studies. RESULTS: The therapeutic application of plant-derived natural inhibitors to treat snakebite envenomation as an auxiliary to antivenom therapy has been gaining significant momentum. Pharmacological reassessment of the natural compounds derived from traditional medicinal plants has demonstrated inhibition of the principal toxic enzymes of snake venoms at various extents to curb the lethal and/or deleterious effects of venomous snakebite. Nevertheless, such molecules are yet to be commercialized for clinical application in the treatment of snakebite. There are many obstacles in the marketability of the plant-derived natural products as snake envenomation antidote and strategies must be explored for the translation of these compounds from drug candidates to their clinical application. CONCLUSION: In order to minimize the adverse implications of snake envenomation, strategies must be developed for the smooth transition of these plant-derived small molecule inhibitors from bench to bedside. In this article we have presented an inclusive review and have critically analysed natural products for their therapeutic potential against snake envenomation, and have proposed a road map for use of natural products as antidote against snakebite.

Advances in the Therapeutic Application of Small-Molecule Inhibitors and Repurposed Drugs against Snakebite.

The World Health Organization has declared snakebite as a neglected tropical disease. Antivenom administration is the sole therapy against venomous snakebite; however, several limitations of this therapy reinforce the dire need for an alternative and/or additional treatment against envenomation. Inhibitors against snake venoms have been explored from natural resources and are synthesized in the laboratory; however, repurposing of small-molecule therapeutics (SMTs) against the principal toxins of snake venoms to inhibit their lethality and/or obnoxious effect of envenomation has been garnering greater attention owing to their established pharmacokinetic properties, low-risk attributes, cost-effectiveness, ease of administration, and storage stability. Nevertheless, SMTs are yet to be approved and commercialized for snakebite treatment. Therefore, we have systematically reviewed and critically analyzed the scenario of small synthetic inhibitors and repurposed drugs against snake envenomation from 2005 to date and proposed novel approaches and commercialization strategies for the development of efficacious therapies against snake envenomation.

Discovery of a multi-target compound for estrogen receptor-positive (ER+) breast cancer: Involvement of aromatase and ERs.

Despite intense research, breast cancer remains the leading cause of cancer-related death in women worldwide, being estrogen receptor-positive (ER+) the most common subtype. Nowadays, aromatase inhibitors (AIs), the selective estrogen receptor modulator (SERM) tamoxifen and the selective estrogen receptor down-regulator (SERD) fulvestrant are used as therapeutic options for ER+ breast cancer, since they interfere directly with the production of estrogens and with the activation of estrogen-dependent signaling pathways. Despite the success of these treatments, the occurrence of resistance limits their clinical efficacy, demanding the development of novel therapies. Recently, multi-target compounds emerged as promising therapeutic strategies for ER+ breast cancer, as they can potentially modulate several important targets simultaneously. In line with this, in this work, the anti-cancer properties and multi-target action of 1,1-Bis(4-hydroxyphenyl)-2-phenylbut-1-ene, tamoxifen bisphenol (1,1-BHPE), were evaluated in an ER+ breast cancer cell model (MCF-7aro cells). Molecular docking analysis predicted that 1,1-BHPE was able to bind to aromatase, ERα and ERß. In vitro studies showed that, although it did not present anti-aromatase activity, 1,1-BHPE reduced aromatase protein levels and interfered with ERα and ERß signaling pathways, acting as an ERα antagonist and inducing ERß up-regulation. Through these mechanisms, 1,1-BHPE was able to impair breast cancer growth and induce apoptosis. This represents an important therapeutic advantage because the main players responsible for estrogen production and signaling are modulated by a single compound. To the best of our knowledge, this is the first study describing the anti-cancer properties of 1,1-BHPE as a multi-target compound specific for ER+ breast cancer.

Cannabidiol (CBD) but not tetrahydrocannabinol (THC) dysregulate in vitro decidualization of human endometrial stromal cells by disruption of estrogen signaling.

Decidualization, which comprises proliferation and differentiation of endometrial stromal cells (ESCs), is essential for the establishment of a receptive endometrium and pregnancy to occur. A deregulation of decidualization has been associated with miscarriage, infertility and other pregnancy-related disorders. The role of estradiol (E2) on decidualization has already been shown, since it regulates proliferation of ESCs and expression of progesterone receptor. In this study, we investigated the effects of phytocannabinoids, tetrahydrocannabinol (THC) and cannabidiol (CBD), in proliferation and differentiation of ESCs, as well as, in E2 metabolism/signaling. We found that CBD, but not THC, inhibits ESCs differentiation. We also show that CBD prevents the increase on transcript levels of CYP19A1 gene and the elevation of E2 levels that are observed in differentiating ESCs. Moreover, we found that CBD presents anti-aromatase activity. In overall, we highlight a novel effect of CBD on human endometrial differentiation, which may lead to infertility problems.

Conformational diversity induces nanosecond-timescale chemical disorder in the HIV-1 protease reaction pathway.

The role of conformational diversity in enzyme catalysis has been a matter of analysis in recent studies. Pre-organization of the active site has been pointed out as the major source for enzymes' catalytic power. Following this line of thought, it is becoming clear that specific, instantaneous, non-rare enzyme conformations that make the active site perfectly pre-organized for the reaction lead to the lowest activation barriers that mostly contribute to the macroscopically observed reaction rate. The present work is focused on exploring the relationship between structure and catalysis in HIV-1 protease (PR) with an adiabatic mapping method, starting from different initial structures, collected from a classical MD simulation. The first, rate-limiting step of the HIV-1 PR catalytic mechanism was studied with the ONIOM QM/MM methodology (B3LYP/6-31G(d):ff99SB), with activation and reaction energies calculated at the M06-2X/6-311++G(2d,2p):ff99SB level of theory, in 19 different enzyme:substrate conformations. The results showed that the instantaneous enzyme conformations have two independent consequences on the enzyme's chemistry: they influence the barrier height, something also observed in the past in other enzymes, and they also influence the specific reaction pathway, which is something unusual and unexpected, challenging the "one enzyme-one substrate-one reaction mechanism" paradigm. Two different reaction mechanisms, with similar reactant probabilities and barrier heights, lead to the same gem-diol intermediate. Subtle nanosecond-timescale rearrangements in the active site hydrogen bonding network were shown to determine which reaction the enzyme follows. We named this phenomenon chemical disorder. The results make us realize the unexpected mechanistic consequences of conformational diversity in enzymatic reactivity.

The Catalytic Mechanism of Human Transketolase.

We have computationally determined the catalytic mechanism of human transketolase (hTK) using a cluster model approach and density functional theory calculations. We were able to determine all the relevant structures, bringing solid evidences to the proposed experimental mechanism, and to add important detail to the structure of the transition states and the energy profile associated with catalysis. Furthermore, we have established the existence of a crucial intermediate of the catalytic cycle, in agreement with experiments. The calculated data brought new insights to hTK's catalytic mechanism, providing free-energy values for the chemical reaction, as well as adding atomistic detail to the experimental mechanism.

Anandamide targets aromatase: A breakthrough on human decidualization.

In each menstrual cycle endometrial stromal cells (hESC) proliferate and differentiate into specialized decidual cells, a process termed decidualization, which regulates endometrial receptivity. Decidualization is mainly controlled by sex ovarian hormones, estradiol (E2) and progesterone. E2 plays an important role in the expression of the progesterone receptor and promotes the endometrial stromal cells differentiation. Our group previously reported that anandamide (AEA) impairs decidualization through cannabinoid receptor 1 (CB1). In this study, we hypothesized whether AEA inhibitory effect on cell decidualization could be mediated through interaction with aromatase and consequent interference in estradiol production/signaling. We used an immortalized human endometrial stromal cell line (St-T1b) and human decidual fibroblasts (HdF) derived from human term placenta. In cells exposed to a differentiation stimulus, AEA-treatment prevents the increase of the expression of CYP19A1 gene encoding aromatase, E2 levels and of estradiol receptor expression, that are observed in differentiating cells. Regarding CYP19A1 mRNA levels, the effect was partially reverted by a CB1 receptor antagonist and by a COX2 inhibitor. In addition, we report that AEA presents anti-aromatase activity in placental microsomes, the nature of the inhibition being the uncommon mixed type as revealed by the kinetic studies. Structural analysis of the AEA-Aromatase complexes determined that AEA may bind to the active site pocket of the enzyme. In overall we report that AEA inhibits aromatase activity and may affect E2 signaling crucial for the decidualization process, indicating that a deregulation of the endocannabinoid system may be implicated in endometrial dysfunction and in fertility/infertility disorders.

Benchmark of Density Functionals for the Calculation of the Redox Potential of Fe3+/Fe2+ Within Protein Coordination Shells.

Iron is a very important transition metal often found in proteins. In enzymes specifically, it is often found at the core of reaction mechanisms, participating in the reaction cycle, more often than not in oxidation/reduction reactions, where it cycles between its most common Fe(III)/Fe(II) oxidation states. QM and QM/MM computational methods that study these catalytic reaction mechanisms mostly use density functional theory (DFT) to describe the chemical transformations. Unfortunately, density functional is known to be plagued by system-specific and property-specific inaccuracies that cast a shadow of uncertainty over the results. Here we have modeled 12 iron coordination complexes, using ligands that represent amino acid sidechains, and calculated the accuracy with which the most common density functionals reproduce the redox properties of the iron complexes (specifically the electronic component of the redox potential at 0 K, Δ E elec F e 3 + / F e 2 + ), using the same property calculated with CCSD(T)/CBS as reference for the evaluation. A number of hybrid and hybrid-meta density functionals, generally with a large % of HF exchange (such as BB1K, mPWB1K, and mPW1B95) provided systematically accurate values for Δ E elec F e 3 + / F e 2 + , with MUEs of ~2 kcal/mol. The very popular B3LYP density functional was found to be quite precise as well, with a MUE of 2.51 kcal/mol. Overall, the study provides guidelines to estimate the inaccuracies coming from the density functionals in the study of enzyme reaction mechanisms that involve an iron cofactor, and to choose appropriate density functionals for the study of the same reactions.

Mechanistic insights on the reduction of glutathione disulfide by protein disulfide isomerase.

We explore the enzymatic mechanism of the reduction of glutathione disulfide (GSSG) by the reduced a domain of human protein disulfide isomerase (hPDI) with atomistic resolution. We use classical molecular dynamics and hybrid quantum mechanics/molecular mechanics calculations at the mPW1N/6-311+G(2d,2p):FF99SB//mPW1N/6-31G(d):FF99SB level. The reaction proceeds in two stages: (i) a thiol-disulfide exchange through nucleophilic attack of the Cys53-thiolate to the GSSG-disulfide followed by the deprotonation of Cys56-thiol by Glu47-carboxylate and (ii) a second thiol-disulfide exchange between the Cys56-thiolate and the mixed disulfide intermediate formed in the first step. The Gibbs activation energy for the first stage was 18.7 kcal·mol-1, and for the second stage, it was 7.2 kcal·mol-1, in excellent agreement with the experimental barrier (17.6 kcal·mol-1). Our results also suggest that the catalysis by protein disulfide isomerase (PDI) and thiol-disulfide exchange is mostly enthalpy-driven (entropy changes below 2 kcal·mol-1 at all stages of the reaction). Hydrogen bonds formed between the backbone of His55 and Cys56 and the Cys56-thiol result in an increase in the Gibbs energy barrier of the first thiol-disulfide exchange. The solvent plays a key role in stabilizing the leaving glutathione thiolate formed. This role is not exclusively electrostatic, because an explicit inclusion of several water molecules at the density-functional theory level is a requisite to form the mixed disulfide intermediate. In the intramolecular oxidation of PDI, a transition state is only observed if hydrogen bond donors are nearby the mixed disulfide intermediate, which emphasizes that the thermochemistry of thiol-disulfide exchange in PDI is influenced by the presence of hydrogen bond donors.

Calculation of distribution coefficients in the SAMPL5 challenge from atomic solvation parameters and surface areas.

In the context of SAMPL5, we submitted blind predictions of the cyclohexane/water distribution coefficient (D) for a series of 53 drug-like molecules. Our method is purely empirical and based on the additive contribution of each solute atom to the free energy of solvation in water and in cyclohexane. The contribution of each atom depends on the atom type and on the exposed surface area. Comparatively to similar methods in the literature, we used a very small set of atomic parameters: only 10 for solvation in water and 1 for solvation in cyclohexane. As a result, the method is protected from overfitting and the error in the blind predictions could be reasonably estimated. Moreover, this approach is fast: it takes only 0.5 s to predict the distribution coefficient for all 53 SAMPL5 compounds, allowing its application in virtual screening campaigns. The performance of our approach (submission 49) is modest but satisfactory in view of its efficiency: the root mean square error (RMSE) was 3.3 log D units for the 53 compounds, while the RMSE of the best performing method (using COSMO-RS) was 2.1 (submission 16). Our method is implemented as a Python script available at https://github.com/diogomart/SAMPL5-DC-surface-empirical .

HMG-CoA Reductase inhibitors: an updated review of patents of novel compounds and formulations (2011-2015).

INTRODUCTION: Statins are remarkably safe and efficient medications that are the mainstay of hypercholesterolemia treatment and have proven to be an invaluable tool to lower the risk of acute cardiovascular events. These compounds are inhibitors of 3-hydroxy-methylglutaryl CoA reductase (HMG-R), the rate-limiting enzyme in cholesterol biosynthesis. In spite of their success, they present undesirable side effects and are now loosing patent protection, which provides a great opportunity for the development of new and improved statins. Areas covered: This review summarizes the new patents for HMG-R inhibitors for the 2011-2015 period. Combinations of existing statins with other drugs are also addressed, as well as novel applications of existing statins. Expert opinion: Recent efforts for the discovery of HMG-CoA-R inhibitors has resulted in several new molecules. Most of these are based on commercially available statins, including sterol and terpenoid derivatives. A few peptides have also been patented. However, the origin of the side effects caused by previous statins continues to be, to a large extent, unknown. Although the patents published in the past 5 years are promising, and might result in new drugs, there is still no way to know if they will present reduced toxicity. Only future clinical trials will answer this question.

Reaction Mechanism of Mycobacterium Tuberculosis Glutamine Synthetase Using Quantum Mechanics/Molecular Mechanics Calculations.

This paper is devoted to the understanding of the reaction mechanism of mycobacterium tuberculosis glutamine synthetase (mtGS) with atomic detail, using computational quantum mechanics/molecular mechanics (QM/MM) methods at the ONIOM M06-D3/6-311++G(2d,2p):ff99SB//B3LYP/6-31G(d):ff99SB level of theory. The complete reaction undergoes a three-step mechanism: the spontaneous transfer of phosphate from ATP to glutamate upon ammonium binding (ammonium quickly loses a proton to Asp54), the attack of ammonia on phosphorylated glutamate (yielding protonated glutamine), and the deprotonation of glutamine by the leaving phosphate. This exothermic reaction has an activation free energy of 21.5 kcal mol(-1) , which is consistent with that described for Escherichia coli glutamine synthetase (15-17 kcal mol(-1) ). The participating active site residues have been identified and their role and energy contributions clarified. This study provides an insightful atomic description of the biosynthetic reaction that takes place in this enzyme, opening doors for more accurate studies for developing new anti-tuberculosis therapies.

Divalent metal ion-based catalytic mechanism of the Nudix hydrolase Orf153 (YmfB) from Escherichia coli.

YmfB from Escherichia coli is the Nudix hydrolase involved in the metabolism of thiamine pyrophosphate, an important compound in primary metabolism and a cofactor of many enzymes. In addition, it hydrolyzes (d)NTPs to (d)NMPs and inorganic orthophosphates in a stepwise manner. The structures of YmfB alone and in complex with three sulfates and two manganese ions determined by X-ray crystallography, when compared with the structures of other Nudix hydrolases such as MutT, Ap4Aase and DR1025, provide insight into the unique hydrolysis mechanism of YmfB. Mass-spectrometric analysis confirmed that water attacks the terminal phosphates of GTP and GDP sequentially. Kinetic analysis of binding-site mutants showed that no individual residue is absolutely required for catalytic activity, suggesting that protein residues do not participate in the deprotonation of the attacking water. Thermodynamic integration calculations show that a hydroxyl ion bound to two divalent metal ions attacks the phosphate directly without the help of a nearby catalytic base.

Isomerization of Δ5-androstene-3,17-dione into Δ4-androstene-3,17-dione catalyzed by human glutathione transferase A3-3: a computational study identifies a dual role for glutathione.

Glutathione transferases (GSTs) are important enzymes in the metabolism of electrophilic xenobiotic and endobiotic toxic compounds. In addition, human GST A3-3 also catalyzes the double bond isomerization of Δ5-androstene-3,17-dione (Δ(5)-AD) and Δ(5)-pregnene-3,20-dione (Δ(5)-PD), which are the immediate precursors of testosterone and progesterone. In fact, GST A3-3 is the most efficient human enzyme known to exist in the catalysis of these reactions. In this work, we have used density functional theory (DFT) calculations to propose a refined mechanism for the isomerization of Δ(5)-AD catalyzed by GST A3-3. In this mechanism the glutathione (GSH) thiol and Tyr9 catalyze the proton transfer from the Δ(5)-AD C4 atom to the Δ(5)-AD C6 atom, with a rate limiting activation energy of 15.8 kcal · mol(-1). GSH has a dual function, because it is also responsible for stabilizing the negative charge that is formed in the O3 atom of the enolate intermediate. The catalytic role of Tyr9 depends on significant conformational rearrangements of its side chain. Neither of these contributions to catalysis has been observed before. Residues Phe10, Leu111, Ala 208, and Ala 216 complete the list of the important catalytic residues. The mechanism detailed here is based on the GST A3-3:GSH:Δ(4)-AD crystal structure and is consistent with all available experimental data.

PLP undergoes conformational changes during the course of an enzymatic reaction.

Numerous enzymes, such as the pyridoxal 5'-phosphate (PLP)-dependent enzymes, require cofactors for their activities. Using X-ray crystallography, structural snapshots of the L-serine dehydratase catalytic reaction of a bacterial PLP-dependent enzyme were determined. In the structures, the dihedral angle between the pyridine ring and the Schiff-base linkage of PLP varied from 18° to 52°. It is proposed that the organic cofactor PLP directly catalyzes reactions by active conformational changes, and the novel catalytic mechanism involving the PLP cofactor was confirmed by high-level quantum-mechanical calculations. The conformational change was essential for nucleophilic attack of the substrate on PLP, for concerted proton transfer from the substrate to the protein and for directing carbanion formation of the substrate. Over the whole catalytic cycle, the organic cofactor catalyzes a series of reactions, like the enzyme. The conformational change of the PLP cofactor in catalysis serves as a starting point for identifying the previously unknown catalytic roles of organic cofactors.

Are hot-spots occluded from water?

Protein-protein interactions are the basis of many biological processes and are governed by focused regions with high binding affinities, the warm- and hot-spots. It was proposed that these regions are surrounded by areas with higher packing density leading to solvent exclusion around them - "the O-ring theory." This important inference still lacks sufficient demonstration. We have used Molecular Dynamics (MD) simulations to investigate the validity of the O-ring theory in the context of the conformational flexibility of the proteins, which is critical for function, in general, and for interaction with water, in particular. The MD results were analyzed for a variety of solvent-accessible surface area (SASA) features, radial distribution functions (RDFs), protein-water distances, and water residence times. The measurement of the average solvent-accessible surface area features for the warm- and hot-spots and the null-spots, as well as data for corresponding RDFs, identify distinct properties for these two sets of residues. Warm- and hot-spots are found to be occluded from the solvent. However, it has to be borne in mind that water-mediated interactions have significant power to construct an extensive and strongly bonded interface. We observed that warm- and hot-spots tend to form hydrogen bond (H-bond) networks with water molecules that have an occupancy around 90%. This study provides strong evidence in support of the O-ring theory and the results show that hot-spots are indeed protected from the bulk solvent. Nevertheless, the warm- and hot-spots still make water-mediated contacts, which are also important for protein-protein binding.

Mechanism of glutathione transferase P1-1-catalyzed activation of the prodrug canfosfamide (TLK286, TELCYTA).

Canfosfamide (TLK286, TELCYTA) is a prodrug that upon activation by glutathione transferase P1-1 (GST P1-1) yields an anticancer alkylating agent and a glutathione derivative. The rationale underlying the use of TLK286 in chemotherapy is that tumor cells overexpressing GST P1-1 will be locally exposed to the released alkylating agent with limited collateral toxicity to the surrounding normal tissues. TLK286 has demonstrated clinical effects in phase II and III clinical trials for the treatment of malignancies, such as ovarian cancer, nonsmall cell lung cancer, and breast cancer, as a single agent and in combination with other chemotherapeutic agents. In spite of these promising results, the detailed mechanism of GST P1-1 activation of the prodrug has not been elucidated. Here, we propose a mechanism for the TLK286 activation by GST P1-1 on the basis of density functional theory (DFT) and on potential of mean force (PMF) calculations. A catalytic water molecule is instrumental to the activation by forming a network of intermolecular interactions between the active-site Tyr7 hydroxyl and the sulfone and COO(-) groups of TLK286. The results obtained are consistent with the available experimental kinetic data and provide an atomistic understanding of the TLK286 activation mechanism.

Computational enzymatic catalysis--clarifying enzymatic mechanisms with the help of computers.

Enzymes play a biologically essential role in performing and controlling an important share of the chemical processes occurring in life. However, despite their critical role in nature, attaining a clear understanding of the way an enzyme acts, i.e. its catalytic mechanism, is a cumbersome task that requires the cooperative efforts of a large number of different scientific techniques. Computational methods offer a particularly insightful way to study such mechanisms, always beautifully complementing the information arising from experimental techniques and working as an excellent alternative for assessing the viability of different mechanistic proposals. This review highlights two important computational strategies to study enzymatic catalysis - the cluster modeling approach and the hybrid quantum mechanical/molecular mechanical (QM/MM) method - complemented with a selection of hand-picked examples of our own work.