Hypertonic solution-induced preconditioning reduces inflammation and mortality rate

Ischemia/reperfusion (IR), systemic inflammatory response (SIRS) and sepsis present dysregulated inflammatory response in different diseases, and are the main causes of hospitalization in the US and worldwide [1, 2]. In addition to this, SIRS, IR and sepsis frequently affect surgical patients. These patients’ common cause of death is shock which is non responsive to vascular contractile drugs. Preconditioning is used in some surgical situations in order to protect patients from vascular clamp and consequent ischemia [3‐7]. Several studies have shown the therapeutic potential of hypertonic solutions as treatment of different insults (sepsis, IR, SIRS). Studies with hypertonic sodium chloride solution revealed a prompt blood pressure restoration in experimental severe hemorrhagic shock [8‐12]. Hypertonic solution treatment was effective in reducing mortality rate of endotoxemic rats [13], as well as prevented lung injury by LPS in the experimental model of ARDS [14, 15]. Hypertonic solution has been used in sepsis’ experimental studies and in small clinical studies [9, 16, 17]. In these studies, hypertonic solution improved tissue perfusion and reduced inflammation. On the other hand, hyperosmolarity can be a side effect of hypertonic solution used for therapeutic purposes.

Recent studies showed that hyperosmolarity causes cell stress activating intracellular MAP kinases pathway [18‐20]. As a consequence of osmotic stress there is an activation of a nonspecific inflammatory response with release of cytokines [20]. IL1β in human aortic endothelial cells, IL6 in peritoneal macrophages of rats and IL8 in human peripheral blood mononuclear cells [9, 17, 21]. Small stress with different substances (e.g. LPS or cytokines) repeated several times can induce a protection [22‐24]. Taking in account the fact that osmotic stress induces an inflammatory response through MAPK pathway we hypothesized that small doses of hypertonic solution injected days before a LPS challenge could induce also a tolerance or preconditioning state in mice. The advantage of hypertonic solution is lower toxicity as inducer of tolerance in patients compared to LPS or cytokines. Thus, this strategy can be used in surgical patients at high risk to develop SIRS, ischemia/reperfusion or sepsis.

The aim of this study was to analyze hypertonic saline 7.5% as a potential inducer of preconditioning in mice, protecting animals from a lethal dose of LPS and modulating the systemic inflammatory profile when administrated previously to LPS.

Three different doses of Hypertonic Saline 7,5% (HS) (4, 6 and 8 mL/kg) in alternated days were tested to determine which dose of HS was effective in inducing a preconditioning. In the HS8 group (we used increasing doses 4, 6 and 8 mL/kg to avoid hyperchloremia of animals). A lethal dose of LPS at the 8th day was injected i.p. for the survival study. We observed on Fig. 1a that hypertonic saline pre-conditioning increased animal survival after LPS injection when used increased doses of HS at the final concentration of 8 mL/Kg (p < 0.05). Figure 1b shows results for negative controls comparing 7.5, 3.5 and 0.9% NaCl solutions and also 20% Manitol, the data shows effective reduction in mortality only in 7.5% NaCl solution. Animals injected with PBS buffer instead LPS did not present any death.

Figure 2 shows the results of different solute for hyperosmotic solution on mortality rate after a lethal dose of LPS. The protective effect was related to sorbitol, urea and HS with the same osmotic value. Taking in account our results, we observed that preconditioning effect was osmolarity-dependent. Therefore, we looked at different solutions with the same osmolarity to find out if the preconditioning effect observed by the hypertonic saline solution would be reproducible using other solutions. We observed that all solutions were able to induce a preconditioning state and increased survival rate compared to LPS group (p < 0.05).

Preconditioning process was analyzed by measuring different plasmatic cytokines production. The cytokines peak presents two different patterns: Cytokines that increased their expression after first dose and then maintaining elevated expression until second dose, decreasing after third dose (Fig. 3a, c, e, f, g, i, j, l) and the cytokines that increased their expression after a second dose and decreased their expression after third dose (Fig. 3b, d, h, k).

In order to identify the possible pathway used by the hypertonic solution to induce preconditioning, we performed a micro array assay in the peritoneal macrophages of animals. Gene expression array showed an important change in several genes exposed to hypertonic solution (Table 1). For example, gene related to cytokines receptors were reduced after hypertonic solution i.e. il1r2, and Mmp8. On the other hand, hypertonic solution increased Mapkbp1 and Atf3 that is related to hypertonic activation of cell signaling to preconditioning.

After confirming the potential effect of HS preconditioning in regulating the immune response, we studied whether HS preconditioning was capable to modulate systemic inflammatory response of animals subjected to endotoxemia by intraperitoneal LPS injection (Fig. 4). As expected, HS preconditioning decreased all cytokines production analyzed in animals treated before endotoxemia induction compared to animals without treatment (p < 0.05) (Fig. 4a, c, e, f, g, i, j, l, m). The effect of reducing cytokines production was not related to a decrease of leucocytes or neutrophils in the blood of animals which were induced to preconditioning as shown in Fig. 5. Thus, in accordance of our hypothesis, HS preconditioning avoided systemic inflammation of endotoxemic animals.

Although the benefits of HS treatment in modulate different injuries are stablished in literature, there are no studies showing the potential of HS in vivo developing preconditioning state able to modulate inflammation. Our study provides, for the first time, data showing that repeated IV infusion of Hypertonic Solution (HS) induces a preconditioning state able to protect animals from lethal LPS injection. We showed that HS preconditioning reduced mortality rate by lethal LPS injection as well as reduced the cytokines release in plasma. Serum cytokine time course after each HS dose showed progressive preconditioning development. The mechanistic pathway related to this process involves the MAPK pathway as observed by our micro array results that showed an increasing in Mapkbp and Atf3 expression. It is noteworthy that MAPK is a kinase activated by osmotic stress and phosphorylated ATF3 which inhibit at the gene promoter the production of cytokines [25].

Experimental studies have shown some benefits of hypertonic solution compared to isotonic fluid for resuscitation after trauma and hemorrhagic shock [8, 21]. Hypertonic solutions (HS) have been shown to reduce systemic inflammation and organ failure associated with resuscitated hemorrhagic shock [21]. Here we showed that the pretreatment with increased doses of HS (4, 6 and 8 mL/Kg) was able to induce a preconditioning state in animals that significantly reduced mortality rate of endotoxemic animals. There is one study in vivo using the term preconditioning to HS previous to ischemia/reperfusion insult in mice [26]. However, in this study the HS was infused 1 h before ischemia/reperfusion in the animals, it is worth noting that so close infusion of HS can protect the animals not by a preconditioning effect but due to volume expansion. Other authors have verified in a similar model of ischemia/ reperfusion that HS infusion has more potent effect in different time points [27]. The authors reported that several parameters demonstrate that HS administered pre-reperfusion is more effective than administered pre-ischemia [27]. Preconditioning should present protection at any moment of infusion [21, 28].

The next question in our study was to verify whether preconditioning state was due to osmotic stress or to solute. The question is if this effect is specific to hypertonic saline or any other osmotic stress could result in similar changes [29]. Our results showed that every hyperosmolar solution were able to decrease mortality rate on endotoxemic animals, so preconditioning is dependent on hyperosmotic stress. Negative controls with lower osmotic solutions in special 0.9% NaCl did not show protection as HS. Other studies on macrophage culture report that the effect of hypertonic preconditioning with either NaCl or mannitol is only transient, returning macrophage function within 20 h after hypertonic preconditioning [29]. Our results confirm that preconditioning induce by HS was effective even after an interval of 48 h, in fact the protection last 10 days, that was the complete time of this study.

There are studies reporting that osmotic stress deflagrates intracellular signaling through mitogen-activated protein kinase (MAPK). The components of the MAPK cascades, TAK1 and MEKKs, activate IKK leading to NF-κB activation [30, 31]. In our analysis of gene expression comparing normal saline (NaCl 0.9%) versus hypertonic saline (7.5%) treatment we found several genes that were changed by hypertonic solution. In particular, we highlight the finding that Mapkbp1 increased 166-fold and Atf3 increased 150-fold in animals treated with hypertonic versus saline solution. The literature showed that hormones, nutrient depletion, osmotic shock, oxidative stress, DNA damage activate Mitogen-activated/stress protein kinase (MAPK/SAPK) pathways [25, 32]. MAPK induced the phosphorylation of transcription factors of the ATF/CREB family and regulated the transcription of target genes. Therefore, ATF3 negatively regulate transcription of pro-inflammatory cytokines. Genes that possess ATF/CREB promoter binding sites within close proximity of NF-κB sites, are ATF3 modulate NF-κB related transcription [25, 32]. Since HS induced preconditioning in animals, we hypothesized whether this preconditioning state would be able to protect animals from an acute inflammation developed by LPS injection. Our results showed that animals submitted to HS preconditioning had a decreased cytokines expression compared to animals without treatment, suggesting a therapeutically effect of HS preconditioning in controlling inflammation.

Finally, our study demonstrated for the first time that HS can induce preconditioning in vivo lasting at least 10 days. Preconditioning effect was dependent on osmotic stress, as several hyperosmotic solutions reproduces the protection to lethal LPS. HS cause alterations in inflammatory cells by modulation of different genes involved in osmotic stress (e.g. MAPK) observed by gene expression analysis. The possible mechanistic pathway is the activation of MAPK and ATF3 activation of cascade with final reduction in the production of cytokines.