One of the fascinating aspects of redox biology – and even one of the major challenges to the understanding of these processes – is the fact that relatively poorly complex intermediates from the biochemical point of view can perform elaborate and finely regulated cellular functions. In this sense, thiol-proteins have been increasingly studied as redox adapters capable of translating redox reactions in the regulation of specific molecular targets, performing multiple cellular functions. One of these thiol-proteins, addressed in various ways in the context of RIDC Redoxoma, is the protein disulfide isomerase (PDIA1).

The PDIA1, or simply PDI, is the prototype of the PDI family, which contains more than 20 members, and belongs to the thioredoxin superfamily. PDI, essential for the survival of our cells, is an abundant protein, expressed in almost all mammalian tissues. Its classic function is to catalyze the insertion of disulfide bonds into nascent proteins in the endoplasmic reticulum (ER), guaranteeing its correct folding essential to cellular protein homeostasis. The disulfide bonds, which stabilize proteins, are bonds resulting from the oxidation of two thiol groups. The PDI also exhibits chaperone activity, which per se is not directly redox-dependent.

PDI is located predominantly in the endoplasmic reticulum, but also, in smaller amounts, on the cell surface, extracellular medium, and possibly also in the cytosol and nucleus. Cell-surface/extracellular PDIA1 is the best characterized pool outside the endoplasmic reticulum and regulates events related to virus internalization, thrombosis, platelet activation, and vascular remodeling. The various functions of the PDI occur in parallel to its characteristic structure. It is made up of 508 amino acids arranged as a U-shaped structure, comprising sequential domains named –a–b–b’–a’, plus a C-terminal domain. The “a” domains in each arm of the “U” contain active redox catalytic sites with CGHC motifs (cysteine, glycine, histidine, cysteine), which allow PDI to undergo reduction-oxidation cycles. The “b” domains at the bottom of the “U” are rich in hydrophobic residues involved in recognition and binding to protein substrates.

The group of Professor Francisco R.M. Laurindo, from the Instituto do Coração da Faculdade de Medicina at Universidade de São Paulo, and a member of the RIDC Redoxoma, has shown over the years an involvement of PDI outside the endoplasmic reticulum in several redox signaling pathways apparently independent of their function in protein folding. Such functions include interaction with the NADPH oxidase complex (Nox) and the cell cytoskeleton; in the case of the extracellular pool, the role of PDI in vascular remodeling and mechanotransduction (the process by which cells convert mechanical stimuli into a chemical response). PDI has also been studied in collaborations among the groups that form the Center for Research on Redox Processes in Biomedicine (Redoxoma), a Research, Innovation, and Dissemination Center (RIDC) supported by FAPESP. Taken together, these data have reinforced a relevant role of PDI in the pathophysiology of cardiovascular and neurodegenerative disorders, diabetes and cancer.

Earlier this year, Laurindo’s group published three articles that expand the understanding of PDI’s multiple roles.

The dual role of PDIA1 on Nox1 NADPH oxidase regulation in cancer

The production of oxidants can both stimulate and inhibit tumor growth. Oxidants generate phenomena that give rise to tumors, such as cell growth and migration. But, from a certain point, they are responsible for apoptosis, i.e. programmed cell death, and regression of the tumor. The balance between these two moments is delicate.

It is known that PDI is implicated in the progression of cancer by mechanisms yet poorly understood. Studies show that PDI is overexpressed in melanoma, lymphoma, hepatocellular carcinoma, cancer of the brain, kidney, ovary, prostate, and lung. It is often associated with metastasis, invasiveness, and drug resistance.

The relationship between PDI and the production of oxidants involves the NADPH oxidase (Nox), a family of enzymes that catalyzes the reduction of molecular oxygen generating the superoxide radical anion, which, in turn, participates in the generation of other oxidants. Several previous studies have indicated that PDI is involved in the activation of Nox, particularly the Nox1 subtype, in vascular cells.

Studying colorectal cancer, in which Nox 1 plays an important role, the researchers from Laurindo’s group observed a dual effect of PDIA1 dependent on different degrees of activation of the KRas oncogene. The research was conducted during the Ph.D. of Tiphany Coralie De Bessa and the results were published in the journal Cell Death and Disease.

The dual effect of PDIA1 is related to the level of activation of the Ras protein. The Ras protein is encoded by the Ras proto-oncogene whose family consists of three genes associated with various cellular growth and morphology functions, and which are involved in human carcinogenesis: HRas, KRas, and NRas. KRas is the most mutated gene in colorectal cancer, being a therapeutic target. A characteristic of the cell with KRas activating mutation is that it activates a series of pathways that generate oxidants, including NADPH oxidase.

In cells with moderate levels of KRas activation, PDI supports the production of superoxide dependent on Nox1 activation, similar to what had previously been observed in vascular cells. However, in cells with mutations generating high activation of KRas, PDIA1 limits the production of the oxidant. The mechanism of this behavior involves the associated increase of the activity of Rac1, which is a subunit of Nox1.

In cells with normal or moderate levels of KRas activity, “PDI supports the activation of Nox1 by a route dependent on an interaction with Rac1, which we had previously seen in vascular cells. However, the novelty of this study is that there seems to be a ceiling effect, that is, the PDI, while activating the Nox1, imposes a kind of maximum limit of activation, and this limit seems to be associated to the degree of activation of the Rac1,” Laurindo explains. Rac1 and RhoA belong to the family of small signaling proteins Rho GTPases, which act as molecular switches, particularly in processes related to the cytoskeletal.

When Rac1 is over-activated by a PDI-independent mechanism, in the case by direct activation by KRas, the limit imposed by the PDI is deactivated and the Nox1 remains active so that, in this case, the PDI becomes a suppressor of this process. Among the signaling pathways involved in these processes, the researchers showed association with Stat3 and GSK3-beta proteins.

According to the researchers, PDIA1 may act as a regulatory mechanism for the production of oxidants by Nox1 in tumors. Potentially, suppression of oxidant production by increased expression of PDI in tumors with KRas mutations could be associated with tumor escape mechanisms.

PDI in human plasma: a new marker of endothelial function

Although it is not known exactly how PDI reaches the surface of the cell and is secreted, effects of the extracellular pool of that protein on thrombosis and vascular remodeling are known. But is there, in fact, a pool of circulating plasma PDI in humans?

This was the first question that guided the research conducted during the doctorate of Percíllia Victória S. de Oliveira, whose results were published in the journal Redox Biology. Previously, Oliveira and Laurindo authored an extensive review on the implications for diseases of the circulating thiol/disulfide pool in plasma.

To answer the question, the researchers used and validated an ELISA assay for specific detection of PDI in the plasma of 35 healthy volunteers. They detected and quantified the circulating protein, and found that the variability in PDI concentrations is high among individuals but surprisingly low in the same individual over time and under different conditions.

According to the researchers, this behavior of plasma concentrations indicates that these can be windows to reveal an individual signature pattern of plasma proteins. “Thinking about the role of PDI in system homeostasis, it makes sense to find a constant individual level of the protein, as if it was an indicator of homeostatic phenomena that occur in the vascular wall. We detected picograms of PDI, which indicates that it may not actually have biological effects, but is a marker of other proteins present in the plasma,” Laurindo said.

To investigate this hypothesis, human plasma samples were classified into two groups: PDI-rich plasma and PDI-poor plasma, based on the median of 330 pg/ml, and subjected to proteomic analysis.

Based on proteomics studies, the researchers identified upregulated proteins associated with thrombosis, immune phenomena, and inflammation in PDI-poor plasmas. In PDI-rich plasma, they found upregulated proteins associated with the cytoskeleton and cell maintenance phenomena. No correlations were found between levels of PDI and traditional risk factors, such as cholesterol, triglycerides, and C-reactive proteins, as well as platelet function.

Going further, the investigators have shown that such protein signatures correlate with endothelial function by replicating the results in vitro by incubating cultured endothelial cells with the respective plasmas. “It was interesting because we were able to reproduce a gene expression profile in these cells quite in line with what we had observed in plasma proteome data,” said the researcher. Genes related to apoptosis, immune response, and blood coagulation were upregulated in cells with PDI-poor plasma and downregulated in those incubated with PDI-rich plasma. Also, the proteomic signatures of the secretome of the cells incubated with the different plasmas were concordant with the corresponding plasma signatures.

The next step was to investigate whether such signatures translated into functional responses. Researchers have found that PDI-poor plasma promotes the compromise of endothelial adhesion to fibronectin, an extracellular matrix protein, and a disturbed pattern of cell migration, which results in less ability to repair an endothelial lesion.

An interesting finding, obtained from the evaluation of the PDI levels of plasma samples collected in a validated database of a population composed of individuals with a clinically evident vascular disease, is that patients with cardiovascular events have lower levels of PDI in relation to healthy individuals.

“It is important to stress that we are not saying that we have discovered a new biomarker. Our work is essentially conceptual research, being the first study to point to PDI as a reporter of a specific plasma proteomic signature. In some ways, it has been surprising to find that there are proteomic signatures in apparently healthy individuals independently of other known cardiovascular risk variables, with direct plasma biological effects and associated endothelial responses.”

pecPDI, mecanoadaptation and drunk cells

It was Laurindo’s group that proposed the designation pecPDI for extracellular PDI, which can be secreted (peri) or be on the cell surface (epicellular) – PDI peri/epicellular, pecPDI. In his doctorate, researcher Leonardo Y. Tanaka showed the involvement of pecPDI in vascular remodeling and investigated the protein role in the mechanoresponse, which is an adaptive response to a mechanical force involving the cellular cytoskeleton.

But is PDI a nonspecific inflammatory mediator, or does it have a direct effect on the cellular organization, i.e. acting on the cytoskeleton? Tanaka advanced this study is his postdoc by analyzing various models of cytoskeletal remodeling and showing that in all of them pecPDI plays an important role in regulating the fine organization of the cytoskeleton and the various processes that depend on it.

The results of this research were published in the American Journal of Physiology-Heart and Circulatory Physiology, and Leonardo Y. Tanaka is the corresponding author of the article.

The cytoskeleton is responsible for maintaining cells structure and internal organization, allowing them to perform essential functions such as cell division and movement.

One of the approaches used in the study was the cyclic mechanical distension, which imitates the forces suffered by the vessels in the body. These experiments allowed the researchers to note, in vascular smooth muscle cells, that the neutralization of pecPDI disorganizes the cytoskeleton and significantly disrupts cell repositioning. Also, in collaboration with researchers from the Physics Institute of USP, they used traction microscopy to show that pecPDIA1 organizes the distribution of intracellular force.

“Cells in which PDI is inhibited are unable to establish the contractile moment, which correlates with a worsened ability to migrate. The cell in which the total PDI was inhibited by gene silencing completely loses the capacity of ordered movement, that is, it does not migrate at all. The cell in which only the pecPDI is inhibited is able to travel by a normal distance but has the persistence property altered. That is, it can not persist in the same direction, it moves as if it was drunk, it loses the fine regulation of cellular targeting during the migration,” explained Laurindo.

The researchers investigated two major mechanisms of pecPDI action in the cellular cytoskeleton organization. One is the protein interaction with integrins, which are adhesion molecules that bind the cytoskeleton to the extracellular matrix and are regulated by the oxidation of their thiols by PDI. And another would be the activation of Rhoa at specific cell sites.

Among the proteins that act on the cytoskeleton are the RhoGTPases. To migrate, the cell depends on the temporal and spatial coordination of several processes that allow it to move efficiently. To move, the cell makes a coordinated series of adhesions and releases, alternately in different places: that is, it adheres in the front and releases behind, then it adheres back and releases in the front, to be able to stretch, and repeats the process. This balance is regulated by RhoGDI, which governs the sequence and site of activation of Rac1 and RhoA. The researchers demonstrated, using a fluorescent sensor of RhoA activation, that pecPDI plays a role in regulating RhoA polarization.

“RhoA and Rac1 proteins are a kind of yin and yang. At the time and place when one of them is high, the other is low. They produce opposite effects and this is very important for the cell to be able to migrate and to have a migration direction,” Laurindo said.

The results obtained reinforce the idea that pecPDI is a fine cytoskeletal organizer and the main pool responsible for the persistence of cell migration.

Taken together, these three studies add to other contributions of RIDC Redoxoma indicating new functions and PDI new mechanistic pathways. According to Laurindo, the PDI is increasingly an investigative link of the group and a new frontier for the understanding of how redox pathways can perform finely regulated cellular functions.

The article Subverted regulation of Nox1 NADPH oxidase-dependent oxidant generation by protein disulfide isomerase A1 in colon carcinoma cells with overactivated KRas, by Tiphany Coralie De Bessa, Alessandra Pagano, Ana Iochabel Soares Moretti, Percillia Victoria Santos Oliveira, Samir Andrade Mendonça, Herve Kovacic, and Francisco Rafael Martins Laurindo, can be accessed at https://www.nature.com/articles/s41419-019-1402-y

The article Protein disulfide isomerase plasma levels in healthy humans reveal proteomic signatures involved in contrasting endothelial phenotypes, by Percíllia Victória S. de Oliveira, Sheila Garcia-Rosa, Ana Teresa A. Sachetto, Ana Iochabel S. Moretti, Victor Debbas, Tiphany C. De Bessa, Nathalia T. Silva, Alexandre da C. Pereira, Daniel Martins-de-Souza, Marcelo L. Santoro, and Francisco RM Laurindo, can be accessed at https://www.sciencedirect.com/science/article/pii/S2213231719300217?dgcid=rss_sd_all

The article Peri/epicellular protein disulfide isomerase-A1 acts as an upstream organizer of cytoskeletal mechanoadaptation in vascular smooth muscle cells, by Leonardo Y. Tanaka, Thaís LS Araujo, Andres I. Rodriguez, Mariana S. Ferraz, Vitor B. Pelegati, Mauro Morais, Aline M. dos Santos, Carlos L. Cesar, Alexandre F. Ramos, Adriano M. Alencar, and Francisco RM Laurindo, can be accessed at https://www.physiology.org/doi/abs/10.1152/ajpheart.00379.2018