The effect of thiamine deficiency on inflammation, oxidative stress and cellular migration in an experimental model of sepsis

Several clinical conditions, such as trauma, burn, severe acute illness and major postoperative surgery generate a series of organic response reactions in response to these injuries, called the systemic inflammatory response syndrome (SIRS). When SIRS is associated with an infection, it is defined as sepsis [1‐3]. Epidemiological studies worldwide [4‐6] and in Brazil [7, 8] indicate that sepsis is a prevalent condition with a high mortality rate that is associated with the emergence of multidrug-resistant pathogens [9, 10] while also generating high social and financial costs [11]. Thiamine (vitamin B1) is a water soluble vitamin that does not accumulate in the body; therefore, it has to be ingested daily [12]. Thiamine pyrophosphate (TPP; the biologically active form of vitamin B1) is an important mitochondrial enzyme cofactor [12] with a key role in energy balance. In addition, TPP participates as a transketolase cofactor in the pentose phosphate pathway, a pathway that replenishes NADPH, which is necessary for the recovery of the reduced form of glutathione (GSH) from its oxidized form. TPP is also a cofactor of the α-ketoglutarate dehydrogenase complex [12], theoretically participating in the recovery of cellular reducing power [13].

The prevalence of TD in critically ill patients has been described [14‐16] and was associated with increased morbidity and mortality [16]. Recent studies have demonstrated that additional attention is needed in the identification of thiamine deficiency (TD) and that thiamine supplementation is necessary not only in intensive care unit patients but also in patients with heart failure [15, 17, 18].

Evidence that TD may cause heart failure has been demonstrated [12, 19, 20], whereby an increase in oxidative stress due to TD led to the apoptosis of cardiomyocytes [21] and produced cardiac remodeling [20, 22]. Structural remodeling and the destruction of cardiomyocyte mitochondria were also observed with TD [20]. A similar finding was observed in a study with human neuroblastoma cells cultured under TD conditions [23].

Several neurological disorders have been associated with TD, including acute confusion and changes in eye movement and gait (Wernicke’s encephalopathy). These findings may be associated with chronic memory impairment (Korsakoff’s syndrome) [12, 24, 25]. Thiamine deficiency may also be associated with brain degenerative conditions such as Parkinson’s disease and Alzheimer’s disease [24‐27]. Thiamine deficiency is also associated with lesions of the endoplasmic reticulum of neurons [28]. The influence of TPP in an experimental model of sepsis in dogs has recently been performed [29], and the following outcomes were analyzed: blood pH, oxygen consumption, base excess, mean arterial pressure and the cardiac index. The authors observed recovery of all of these parameters in the animals treated with TPP, suggesting the involvement of thiamine as a possible protective agent against the onset of sepsis.

Sepsis is associated with an intense inflammatory response [30]. In response to an injury, the initial inflammatory response is the activation of the innate immunity pathways. The innate inflammatory response is a series of “primitive” non-specific reactions that occur in the early hours after the interaction with a pathogen and involve acute phase proteins, the complement system, dendritic cells, macrophages and natural killer cells (NK), with the end result being inflammation [31‐34]. When an activated macrophage comes into contact with a pathogen, it begins to produce TNF-α, which is responsible for an increase in vascular permeability. In addition, pathogen stimulation of macrophages leads to the production of IL-1β and IL-6, which have decisive roles in the genesis of fever by acting on the central hypothalamic temperature control system [34]. The chemotactic factors KC and MCP-1 are involved in the recruitment of neutrophils and monocytes, respectively, and are involved in the perpetuation of the cellular response to injury. In parallel, IL-10 secretion is a self-regulatory mechanism of the inflammatory response, acting as an anti-inflammatory cytokine [30, 31].

The cellular injury observed in sepsis produces highly reactive biochemical intermediates from nitrogen and oxygen [35]. Following an injury, we now know that the increased production of reactive oxygen species (ROS) and the increased activity of clearance mechanisms do not necessarily indicate cell damage [35, 36]. ROS have the ability to react with proteins, lipids, and genetic material. The products of these reactions are known today and include 4-hydroxy 2-nonenal (4-HNE), an aldehyde that results from the lipid peroxidation of the cell membrane [37]. 4-HNE plays a key role in cell signaling and redox balance [38, 39], but in high concentrations, it can be harmful to cells [37‐39].

Experimental sepsis models are necessary to evaluate some aspects of this clinical condition, due to ethical constraints [40‐42]. Many methods are used to simulate hemodynamic, immunological and clinical sepsis events, including surgical manipulation of organs (such as ligation or perforation of the bowel loops) [42] and injection of bacteria or their components (e.g., endotoxins, such as gram-negative lipopolysaccharides) into the bloodstream [41], organs and anatomical cavities. These methods are not free of criticisms from both a methodological standpoint [43, 44] and from technical and ethical standpoints, with regard to animal handling [45]. These experimental models, however, represent important tools for sepsis investigation.

This study analyzed the possible effect of thiamine deficiency on changes in sepsis markers of inflammation (cytokines), the balance of the oxidation-reduction system (based on the presence of a marker of lipid peroxidation), and the leukocyte recruitment profile. The hypothesis in question was that thiamine deficiency in experimental sepsis conditions could result in a pro-inflammatory profile, with greater oxidative stress and cellular migration.

This study was approved by the ethics committee for the use of animals for research at the Roberto Alcântara Gomes Biology Institute – UERJ (CEUA/036/2011).

In this study, we found that mice fed the thiamine deficient AIN93G chow had significantly reduced levels of the biologically active form of thiamine in their serum, providing an experimental condition to assess the effect of thiamine deficiency during sepsis.

First, the normal blood TPP concentration and the number of days necessary for TD induction with TD chow in the mice were determined. Of note, this information had not yet been described in previous research.

The chow intake and mass gain were progressive in the AIN93G complete chow fed animals. In contrast, a decrease in the chow intake and a loss of body weight throughout the study were observed in the mice fed the thiamine deficient chow. This was a profile similar to what was previously described with Wistar rats [20].

Although the animals fed with the thiamine deficient chow for 10 days displayed a significant reduction of TPP in the blood, the decision to carry out the sepsis experiments after 15 days was made to leave no doubt that thiamine deficiency induction had occurred. Because there was no significant difference in the body mass gain and chow intake between the groups on the 15th day, it was not necessary to perform a paired feeding to avoid errors related to ingestion of different amounts of calories, proteins and other nutrients among the groups.

The levels of IL-6, MCP-1 and KC in the blood and of IL-1β, IL-6 and KC in the peritoneum were not different among the septic groups, independent of their thiamine status. In addition, CLP increased the number of bacteria in the peritoneum, except in the TD group, where the bacterial recovery was similar to that observed in the sham groups.

The level of TNF-α in the peritoneal fluid of the septic animals fed with TD chow was significantly greater than that observed in the other groups. TNF-α is a cytokine of the early innate inflammatory response. It is produced within the first hour of injury, with a peak between the second and third hours following injury. For this reason, it is not usually used in the monitoring of septic patients [51] because the timing of the peak TNF-α level may have already passed when these patients are first evaluated. However, TNF-α is an early marker of inflammation commonly used in sepsis models [47]. TNF-α is a marker of the activity of NK cells and macrophages and may help explain the increased bacterial clearance in the peritoneal fluid that was observed in the CLP mice fed the TD chow used in this study. In addition, we observed a persistent high level of TNF-α in the peritoneal fluid of the CLP mice fed the TD chow, although the number of total leukocytes, monocytes and neutrophils in the peritoneal fluid was similar between the septic animals, independent of their thiamine status. This finding suggests that TD may not change cellular recruitment into the peritoneal cavity following sepsis and that another mechanism may be involved in the increased peritoneal bacterial killing observed in the CLP group fed with TD chow, possibly mediated by TNF-α.

Lipid peroxidation is one of the hallmarks of the oxidative stress. Lipid peroxides are usually decomposed into reactive aldehydes such as malondialdehyde (MDA) and 4-hydroxy-2-nonenal (4-HNE), which are also reactive oxygen species. Due to its electrophilic properties, the aldehyde 4-HNE forms adducts with cellular proteins [52]. The present study also demonstrated a greater level of liver proteins modified by 4-HNE in the animals subjected to TD chow compared with what was observed in the other groups. This finding was expected, considering that thiamine may be associated with a lower efficiency of cellular reducing power, since enzymes such as pyruvate dehydrogenase and α-KGDH are responsible for increasing NADH levels in the cell, favoring oxidative stress. Thus, one could suppose that thiamine deficiency also creates greater oxidative stress in the peritoneal cavity, therefore potentially inhibiting bacterial proliferation. The increased bacterial clearance that was observed in this study might initially suggest a better outcome against sepsis in these animals; however, the increased inflammation, as was observed with the increased TNF-α levels, may also be detrimental to sepsis outcomes. Indeed, it was reported that following CLP, an increased mortality rate occurred in animals that displayed an unbalanced inflammatory response, despite a greater bacterial clearance [53].

Extending the analysis of inflammation beyond the first 24 hours after the initiation of sepsis may present a more complete picture of the dynamics of the interleukins.

It is important to remember that the model of sepsis used in the present study can be adjusted to induce a low mortality rate, which may also allow for the investigation of thiamine deficiency in the mechanisms of immunologic tolerance.

TD was associated with a greater bacterial clearance in the peritoneal fluid, greater oxidative stress and a change in the immune response profile in an experimental model of abdominal sepsis in mice. Further investigation is needed to define the impact of these changes on the severity of sepsis, septic mortality, and also to elucidate the participation of other cellular and biochemical mechanisms.