A single dose of eHSP72 attenuates sepsis severity in mice

High levels of extracellular 72 kDa heat shock protein (eHSP72) can be detected in the serum of septic patients and are associated with increased oxidative profiles and elevated rates of mortality among these patients. However, a possible immunomodulatory role for this protein, resulting in tissue protection during sepsis, has never been assessed. In this study, we investigated whether eHSP72 administration could attenuate the severity of sepsis in a mouse peritonitis model. Animals (90-day-old male C57BL/6J mice) were divided into Sepsis (n = 8) and Sepsis + eHSP72 (n = 9) groups, which both received injections of 20% fecal solution [1 mg/g body weight (wt), intraperitoneal (i.p.)], to trigger peritonitis induced-sepsis, whereas a Control group (n = 7) received a saline injection. eHSP72 was administered (1.33 ng/g body wt) to the Sepsis+eHSP72 group, 12 h after sepsis induction. All animals were evaluated for murine sepsis score (MSS), hemogram, core temperature, and glycemia (before and 4, 12, and 24 h after sepsis induction). Treatment with eHSP72 promoted reduced sepsis severity 24 h after sepsis induction, based on MSS scores (Control = 1.14 ± 1.02; Sepsis = 11.07 ± 7.24, and Sepsis + eHSP72 = 5.62 ± 1.72, P < 0.001) and core temperatures (°C; Control = 37.48 ± 0.58; Sepsis = 35.17 ± 2.88, and Sepsis + eHSP72 = 36.94 ± 2.02; P = 0.006). eHSP72 treatment also limited the oxidative profile and respiratory dysfunction in mice with sepsis. Although sepsis modified glycemic levels and white and red blood cell counts, these variables were not influenced by eHSP72 treatment (P > 0.05). Finally, eHSP72 improved the survival rate after sepsis (P = 0.0371). Together, our results indicated that eHSP72 may ameliorate sepsis severity and possibly improve some sepsis indices in mice.

In white blood cells obtained from patients, Briassoulis et al. 21 demonstrated that iHSP72 protein synthesis was activated during sepsis, as iHSP72 protein expression was increased in septic patients compared with that in healthy individuals. However, when septic patients were compared with systemic inflammatory response syndrome (SIRS) patients, lower iHSP72 levels were observed in monocytes 10,11 and neutrophils from septic patients than in SIRS patients 10 . Similarly, neutrophil iHSP72 levels were lower in septic shock patients than in non-septic shock patients 8 .
Because enhanced iHSP72 levels may be associated with improved host defense, the failure of this cytoprotective heat shock response in immune cells may be associated with cytokine burst signaling observed during sepsis 22,23 . Reduced iHSP72 expression in monocytes in response to a lipopolysaccharide (LPS) insult was associated with a cytokine burst 21 . Furthermore, repressed iHSP72 expression has been associated with an increased mortality rate and the repressed expression of CD14/HLA-DR in monocyte and polymorphonuclear cells during sepsis 23 . Additionally, iHSP72 may counteract nuclear factor (NF)-κB activation and the associated burst of pro-inflammatory cytokine signalling 24 . Because eHSP72 may be internalized by antigen-presenting (and other) cells 25 , eHSP72 may be able to act as an authentic iHSP72 and induce a similar protective response.
Therefore, eHSP72 may represent a beneficial factor that can re-establish immunoinflammatory balance during sepsis, which has never before been assessed, although the administration of recombinant eHSP72 has been found to increase survival rates in septic rats 6,26 . Therefore, we aimed to investigate whether the intravenous administration of eHSP72 might attenuate the severity of sepsis in a mouse peritonitis model.

Materials and Methods
Animals and ethics. Male, C57BL/6J mice (n = 24), aged 3-5-months, were obtained from the Regional University of the Northwestern Rio Grande do Sul State (UNIJUÍ) Animal Facility and maintained in a controlled-temperature environment (24 ± 2 °C), with a 12-h light-dark cycle (light from 7:00 a.m. to 7:00 p.m.). The animals received water and food ad libitum. The investigation followed all ethical rules established by Arouca's Act (Brazilian Federal Law no. 11794/2008) and the 8th Edition of the Guide for Care and Use of Experimental Animals, published by the National Research Council of the National Academies (2011; available at https:// grants.nih.gov/grants/olaw/guide-for-the-care-and-use-of-laboratory-animals.pdf), in accordance with Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. All the procedures were reviewed and approved by the Committee of Animal Welfare of the Regional University of the Northwestern Rio Grande do Sul State (UNIJUÍ; protocol CEUA-UNIJUI #048/2016), which adheres to the guidelines of The Brazilian National Council for the Control of Animal Experimentation (CONCEA).
Experimental design. Mice (n = 24) were randomly divided into three groups: Control (n = 7), Sepsis (n = 8) and Sepsis + eHSP72 (n = 9). Animals from Sepsis and Sepsis + eHSP72 groups received an intraperitoneal (i.p.) injection of 20% fecal solution [1 mg/g body weight (wt), containing a bacterial load of approximately 43 and 99 CFUs/g of Escherichia coli and Staphylococcus aureus, respectively, as detailed in Supplementary Material 1] to trigger peritonitis induced-sepsis, whereas the Control group received 0.9% saline (i.p.) 27 . Then, 12 h after sepsis induction, eHSP72 (hspa1a, mouse recombinant low-endotoxin, Enzo ADI-ESP-502, 98.4% homologous with Rattus norvegicus hspa1a) was administered intravenously (i.v.) (1.33 ng/g) to mice in the Sepsis + eHSP72 group. We choose this dose because mice with 30 g of body weight contain approximately 2 ml of blood 28,29 ; thus the concentration of eHSP72 in the blood was expected to reach approximately 20 ng/ml, which was the concentration reported for an experimental mouse model of sepsis 24 h after cecal ligation puncture (CLP) 30 . We administered eHSP72 12 h after sepsis induction, in an attempt to anticipate the immune-mediated responses of eHSP72. All animals were evaluated by recording the murine sepsis score (MSS), glycemia, hemogram, and rectal temperature at the following times: just before group assignment (time 0 h), 4, and 12 h after fecal solution administration for sepsis induction (before eHSP72 administration), and 24 h after sepsis induction. No animal deaths were recorded up to this time point, but the mice were followed up to 48 h after sepsis induction, for the assessment of survival rates (Kaplan-Meier).
Murine sepsis score test. The murine sepsis score (MSS) included variables affected by the infection, including spontaneous activity, response to touch and auditory stimuli, posture, respiration rate and quality (labored breathing or gasping), and appearance (the degree of piloerection). For each of these variables, a score between 0 and 4 was given (See details in ref. 31 ). Two independent, blinded researchers, who were previously trained, analyzed the videos.
Bodyweight, glycemia, and core temperature. Individual animal body weights were assessed using a semi-analytical scale, just before group assignments (time 0 h), 4, and 12 h after fecal solution administration for sepsis induction. At the same time points, blood glucose levels were measured, using a Glucometer Optium Xceed (Abbott) (5 µL of tail blood) 32 , and the core temperatures of the animals were assessed using a Minipa Digital Thermometer MT450, equipped with a rectal probe 24 .
Hematological parameters. Blood samples were collected from each animal by caudal puncture (10 µl), just before group assignment (time 0 h), 4, and 12 h after the administration of fecal solution for sepsis induction. were diluted 1:3 with 0.9% saline, containing 1 µl of anticoagulant [Ethylenediaminetetraacetic acid (EDTA)]. Afterward, slides were subjected to hematological staining to obtain leukocyte counts (neutrophils, lymphocytes, and monocytes) 32 , by using panoptic type staining (Newprov). Smears were analyzed by a professional with experience in the field to confirm automated data acquisition 33 . For automatic determination, a Micros 60 hematology analyzer (Horiba) was used, following the manufacturer's recommendations. The following parameters were obtained: hematocrit level, total leukocyte count, absolute leukocyte count, plus platelet count, platelet count, and mean platelet volume (Horiba-User Manual).
Statistical analysis. All data were verified for normality by the Kolmogorov-Smirnov test, and results are expressed as the mean ± standard deviation. Comparisons among groups and time effects were analyzed by two-way analysis of variance (ANOVA, treatment × time), followed by the Bonferroni post hoc test. Data obtained at the end of the study, such as respiratory frequency, oxidative stress parameters, and HSP70 expression, were analyzed by one-way ANOVA, followed by the Tukey post hoc test. All statistical analyses were performed using GraphPad for Windows, version 8.0. The level of significance was set to P < 0.05. Our experimental approach had a 70% power to detect a survival difference of 0.630, with a significance level (alpha) of 0.05 (two-tailed), as calculated in StatMate Software 2.0 (GraphPad). Kaplan-Meier estimates of cumulative survival were generated (GraphPad) and assessed for standard errors and 95% confidence intervals, using the Cox-Mantel rank test and a Chi-squared value, from which the P-values were obtained.

Results
Animals submitted to sepsis showed increased MSS values both 12 h and 24 h after sepsis induction. However, the severity of sepsis evaluated by MSS was lower 24 h after sepsis induction in mice treated with eHSP72 (Fig. 1a). Sepsis also induced a decrease in mouse body temperatures compared with the Control group; however, eHSP72 administration reduced this effect (Fig. 1b). Furthermore, sepsis-induced a decrease in glycemic levels and body weights compared with the Control group, and eHSP72 treatment did not influence these effects (Fig. 1c,d). We did not observe any relative muscle mass alterations, as assessed by comparisons of the gastrocnemius muscle to body weight ratios, among the groups (Fig. 1e).
The behaviors of each variable evaluated to determine the total MSS, which is shown in Fig. 1, can be observed in detail in Fig. 2. Sepsis worsened all measured variables compared with those in the Control group, including the response to stimulus, eye-related aspects, the respiratory rate, the quality of respiration, appearance, levels of consciousness, and motor activity (Fig. 2a-g). The eHSP72-treated group presented better MSS scores than the Sepsis group in response to the stimulus test, eye-related aspects, the respiratory rate, and the quality of respiration (Fig. 2a-d), whereas levels of consciousness and motor activity were not modified ( Fig. 2e-g).
We evaluated the hematological profiles of mice for 24 h. Total leukocyte counts decreased in mice with sepsis, 24 h after fecal administration, which were not affected by eHSP72 treatment (Fig. 3a). Additionally, animals with sepsis presented increased neutrophil counts (Fig. 3b-d) and decreased lymphocyte (Fig. 3e) and monocyte (Fig. 3f) counts. Sepsis also induced decreases in both the absolute (Fig. 4a) and relative (Fig. 4b) platelet counts, without affecting mean platelet volume (Fig. 4c) or hematocrit levels (Fig. 4d). Thus, sepsis promoted and increased the mean platelet volume to platelet count ratio (MPV/PC) (Fig. 4e).
To gain further insight into the inflammatory response during sepsis, we analyzed the equilibrium among blood cells counts. At 12 and 24 h after sepsis induction, an inflammatory response was observed, as indicated by an increase in the neutrophil-to-lymphocyte ratio (NLR) (Fig. 5a), which was associated with an increase in the platelet-to-lymphocyte ratio (PLR) at 24 h (Fig. 5b). We did not observe any alterations in the monocyte-to-lymphocyte ratio (MLR) (Fig. 5c). Treatment with eHSP72 did not influence these responses.
We observed decreased respiratory rates in mice, 24 h after sepsis induction (Fig. 6a), accompanied by increased oxidative profiles in the lungs, as demonstrated by increased lung lipid peroxidation levels (Fig. 6b) and SOD activity levels (Fig. 6c) in the Sepsis group compared with those in the Control group. Interestingly, the Sepsis + eHSP72 group showed attenuated effects among these parameters and decreased HSP70 levels were observed in the eHSP72-treated group relative to those in the Sepsis group (Fig. 6d).
www.nature.com/scientificreports www.nature.com/scientificreports/ Finally, we performed a separate experiment to evaluate the effects of eHSP72 administration on cumulative survival. An increase in the survival rate was observed for the eHSP72 + Sepsis group compared with that for the Sepsis group (P = 0.0371) (Fig. 7). All animals from the Sepsis group died within the first 36 h after sepsis induction, whereas, during the same time period, the survival rate in the Sepsis+eHSP72 group was 90%. However, 70% of the mice in the Sepsis+eHSP72 group died within 42 h after sepsis induction (Fig. 7).

Figure 1.
Effects of eHSP72 administration on murine sepsis score (MSS), rectal temperature, glycemia, and biometric profile, in mice with sepsis. Sepsis was induced by the administration of a 20% fecal solution (1 mg/g, i.p.), whereas the Control group (n = 7) received saline. eHSP72 was administered (1.33 ng/g) to the Sepsis + eHSP72 group,12 h after sepsis induction (n = 9), whereas the Sepsis group received saline (n = 8). MSS (a), rectal temperature (b), glycemia (c), loss of body weight (d), and muscle mass (e). A two-way ANOVA, followed by Bonferroni post hoc test, was used in A, B, C, and D. One-way ANOVA, followed by the Tukey test, was used in E. * P < 0.05 vs. Control; ** P < 0.05 Sepsis vs. Sepsis + eHSP72.

Figure 2.
Effects of eHSP72 administration on individual components of the murine sepsis score (MSS). Experimental conditions were those described for Fig. 1 Response to a stimulus (a), aspect of eyes (b), respiratory frequency (c), respiratory quality (d), appearance (e), level of consciousness (f), and activity (g). A two-way ANOVA, followed by Bonferroni post hoc test, was used. * P < 0.05 vs. Control; ** P < 0.05 Sepsis vs. Sepsis + eHSP72.

Discussion
In our study, we demonstrated that eHSP72 treatment, when used during the early stages of sepsis, attenuated sepsis severity in an animal model, both clinical and subclinically. eHSP72 administration promoted positive effects on the survival rate, responsiveness, and respiratory functions, which were related to the attenuation of the lung oxidative profile. Although some variables were not improved by eHSP72, none of the variables was worsened by this treatment in the mouse model of sepsis.
We believe that eHSP72 administration may mimic the roles played by alarmins or danger signals (danger-activated molecular patterns or DAMPs) 25,36 , due to its interactions with pattern recognition receptors, such as TLR2, TLR4, lectin-like oxidized low-density lipoprotein receptor-1 (LOX-1), CD91, CD94 (C-type  www.nature.com/scientificreports www.nature.com/scientificreports/ lectin), CD40, and chemokine receptor CCR5, resulting in DAMP-like signaling 25,36,37 . eHSP72 can bind to a wide repertoire of cell-surface receptors that may result in internalization 38,39 ; therefore, eHSP72 administration may induce intracellular responses. Additionally, cell surface molecules of the interrelated scavenger receptor and C-type lectin families may also internalize eHSP72 39 . At least three members of the scavenger receptor Figure 5. Effects of eHSP72 administration on the ratio of blood cells in mice with sepsis. Experimental conditions were those described for Fig. 1. Neutrophil-to-lymphocyte ratio (a), platelets-to-lymphocyte ratio (b) and monocyte-to-lymphocyte ratio (C). Two-way ANOVA, followed by Bonferroni post hoc test was used. * P < 0.05 vs. Control; ** P < 0.05 Sepsis vs. Sepsis + eHSP72. Figure 6. Effects of eHSP72 administration on lung function, oxidative, and cell stress biomarkers in mice with sepsis. Experimental conditions were those described for Fig. 1. Respiratory frequency (a), lung lipid peroxidation levels (b), SOD activity (c), and 70 kDa heat shock protein (HSP70) expression (d). Twoway ANOVA, followed by Bonferroni post hoc test was used. * P < 0.05 vs. Control; ** P < 0.05 Sepsis vs. Sepsis+eHSP72. Representative blots are shown at the bottom.
By binding cell-surface receptors, eHSP72 administration can modify intracellular cell signaling, providing anti-inflammatory benefits. eHSP72 has been demonstrated to reverse inflammatory imbalances through mitogen-activated protein kinase (MAPK) pathways, as described in rodent brain structures 40 . The intrahippocampal infusion of eHSP72 was also shown to enhance MAPK pathways by increasing the activity of extracellular signal-regulated kinase (ERK) and decreasing the activity of protein kinase 38 (p38) and JNK 40 . eHSP72 administration also promotes cardiovascular benefits during sepsis 6 , which may act to preserve cerebral hemodynamics and account for the observed improvements in the MSS evaluations that were observed in our study in eHSP72+Sepsis mice. Sepsis often results in central nervous system dysfunction; therefore, our MSS results, particularly those for the consciousness levels of mice, may indicate the potential neuroprotective function of eHSP72 treatment against infection-related dysfunction 41 . These results agree with the findings reported Chang et al., who demonstrated the amelioration of sepsis-related cardiovascular dysfunction, induced by the direct action of eHSP72 in the ventrolateral rostral amygdala 42 .
The administration of eHSP72 resulted in reduced levels of the anti-inflammatory iHSP70 in the lungs of treated animals. However, the plasma levels of eHSP72 have been shown to return to basal levels only 180 min after the intravenous administration of eHSP72, which suggests the tissue uptake of eHSP72 6 . The heat shock factor-1 (HSF-1)-induced expression of iHSP72 can be blocked by excess iHSP70; therefore, the uptake and internalization of eHSP72 by immune cells 43 may similarly decrease the expression of lung iHSP70 observed in the present study. eHSP72 treatment also attenuated the oxidative profile induced by sepsis, in our study. Our investigation showed that eHSP72 attenuated sepsis-induced alterations in the respiratory pattern and the lung oxidative profile, which are commonly observed during peritonitis 31,44,45 . Because the oxidative burst in the lungs is related to leukocyte infiltration, especially neutrophil activity against microorganism invasion 46 , the ability of eHSP72 to attenuate respiratory dysfunction by decreasing the neutrophil-and monocyte-dependent production of reactive oxygen species (ROS) after exposure to bacterial antigens 47 may explain the respiratory system benefits observed in our study. The reduction of vascular permeability induced by eHSP72 48,49 may also be a plausible explanation for these findings.
In our study, eHSP72 administration attenuated the sepsis-induced decrease in body temperature by 50% (0.5 °C). Because eHSP72 may decrease NO production by leukocytes during infection, which has been associated with improved cardiovascular function in an animal sepsis model 50 , and NO is also involved in the central control of thermoregulation 51 , eHSP72 may affect mouse body temperature through the modulation of NO metabolism in the central nervous system, as previously suggested 16,51 .
In addition to animal model studies, the role of HSP72 (both eHSP72 and iHSP72) has been investigated in humans. Patients with sepsis and severe sepsis demonstrated increased serum eHSP72 compared with controls and with SIRS patients, which enables the discrimination of sepsis from SIRS in patients across all age groups [9][10][11] . eHSP72 levels were also detected at increased levels among non-survivors compared with survivors and are considered to be a strong predictor of mortality 8 . When sepsis patients were compared with SIRS patients, lower iHSP72 levels were found in monocytes 10,11 and neutrophils from sepsis patients 10 . www.nature.com/scientificreports www.nature.com/scientificreports/ The association between eHSP72 and mortality in sepsis appears primarily in conjunction with an oxidative profile 4 . The elevation of eHSP72 during sepsis may represent a tentative immune-related danger signal 25,36 that fails under oxidative stress conditions, during which eHSP72 may lose its functionality, including the ability to trigger immune signaling pathways 20 .
Similarly, when eHSP72 levels increase, iHSP72 levels are repressed in septic patients 52 , and the elevated concentrations of these proteins in the bloodstream may be related to glucocorticoid receptor expression, induced by eHSP72 in monocytes, which facilitate cortisol binding to counterbalance the increased inflammatory response during early sepsis. As discussed by Vardas and colleagues 8 , exogenous steroid support may not be necessary due to signals mediated by the glucocorticoid receptor, associated with stress-activated bio-molecules and organ dysfunction. Together, these pieces of evidence suggest a positive aspect of our study, which was that eHSP72 administration that occurred 12 h after sepsis induction was able to attenuate the severity of related neurological functions, as observed during the MSS test.
However, human studies examining eHSP72 administration during sepsis have been inconclusive. Briassoulis and colleagues compared various sepsis outcomes, which differ greatly between human and animal studies 3 . Clinical studies have shown a low probability of protection (16.7%) and possible relationships with mortality and infections for eHSP72 administration. In contrast, almost all (94.7%) septic animals in both in vivo and in vitro studies showed biochemically, biologically, and clinically protective effects for eHSP72 during sepsis, in agreement with our study. Our study also demonstrated that a single dose of eHSP72 promoted benefits in a mouse model of sepsis. Further studies should be performed to provide a more complete description regarding the dose-response of eHSP72 on the immune system during sepsis and to investigate cytokine levels and the effects of eHSP72 treatment on immune cell activation, including splenocytes and peritoneal macrophages.

Conclusion
Treatment with eHSP72 promoted benefits in the stimulus-response, body temperature control, respiratory pattern, survival rate, and oxidative profile of a mice sepsis model. Together, our results indicate that treatment with eHSP72 may attenuate the severity of sepsis in an animal model, opening a perspective for future studies against sepsis.

Data availability
The data used to support the findings of this study are available from the corresponding authors upon request.