Mycobacterium leprae Hsp65 administration reduces the lifespan of aged high antibody producer mice

Background Aging process may result in immune modifications that lead to disruption of innate and acquired immunity mechanisms that may induce chronic-degenerative events. The heat shock proteins (Hsp), phylogeneticaly conserved among organisms, present as main function the ability of folding and refolding proteins, but they also are associated with chronic-degenerative disorders. Here were evaluated the role of M. leprae native Hsp65 (WT) and its point-mutated (K409A) on survival and anti-DNA and anti-Hsp65 antibody production of aged genetically selected mice for high (HIII) and low (LIII) antibody production; data from 120- and 270-days old mice (named “adult” or “aged”, respectively) were compared. Results WT Hsp65 administration induces reduction in the mean survival time of adult and aged female HIII mice, this effect being stronger in aged individuals. Surprisingly, the native protein administration increased the survival of aged female LIII when compared to K409A and control groups. No survival differences were observed in aged male mice after Hsp65 proteins inoculation. We observed increase in IgG1 anti-Hsp65 in WT and K409A aged HIII female mice groups and no marked changes in the anti-DNA (adult and aged HIII) and anti-Hsp65 IgG1 or IgG2a isotypes production in adult HIII female and aged male mice. LIII male mice presented increased anti-DNA and anti-Hsp65 IgG2a isotype production after WT or K409A injection, and LIII female groups showed no alterations. Conclusions The results revealed that the WT Hsp65 interferes with survival of aged HIII female mice without involvement of a remarkable IgG1 and IgG2a anti-DNA and anti-Hsp65 antibodies production. The deleterious effects of Hsp65 on survival time in aged HIII female mice could be linked to a gender-effect and are in agreement with those previously reported in lupus-prone mice.


Background
Aging is defined as progressive alterations of biological functions, leading to the onset of diseases and reduced ability to respond to external stimuli [1]. Alongside with the physiological aging events, the immunosenescence accumulates potential modifications in immunological functions and its components. The most important changes include the decrease of the absolute number of lymphocytes, alterations of the activation status of T cells, increasing of serum levels of immunoglobulin (mainly IgA and IgG), limitation of the protective response of specific high affinity antibody, amplification of autoantibody production and a switch for a Th2 pattern of cytokine response [2]. The altered processes in advanced age also result in the failure of self and non-self discrimination [3] and disruption of the innate and acquired immunity mechanisms, which may result in chronic-degenerative events and subsequent loss of life quality [4][5][6][7]. Altogether, these modifications lead to an increased vulnerability to infections [8,9], reduced response to vaccines [10], development of tumors [11,12], and autoimmune or inflammatory diseases [13,14]. In addition, disorders related to the abnormal processing, modification, and aggregation of proteins typically linked to biological properties of the heat shock proteins (Hsp) are reported [15,16].
Drastic alterations in physiological responses to stressful events are related to Hsp production [17][18][19]. Hsp are phylogeneticaly conserved molecules among evolutionary scale [20,21] which assist misfolding molecules and control the arising of toxic protein aggregates, supporting the folding and unfolding of polypeptides for degradation by proteolytic machinery [22,23]. Hsp65, the most abundant and immunogenic protein of mycobacteria [24], is considered a toxin and dominant antigen in infectious diseases, capable of induce humoral and cellular immune responses [25][26][27]. Reports evidenced the immunodominant role of the Hsp60 family in infectious processes [28], besides of the role played in inflammatory processes such as arthritis, type I diabetes, multiple sclerosis and atherosclerosis [29][30][31][32]. In the opposite, some studies demonstrate its regulatory function on immune suppression in rheumatoid arthritis [33] and type I diabetes [34].
Previously, our group evaluated the immunomodulatory effects in vivo of M. leprae Hsp65 on genetically homogeneous (NZBxNZW)F 1 hybrid female mice that develop systemic lupus erythematosus (SLE); the results showed that the native protein (WT) aggravates the lupus progression in mice [35]. On the other hand, the K 409 A, a pointmutated Hsp65 [36], revealed a potential in mitigating lupus aggravation in these mice [37]. Hsp65 administration also increased eye lesions in mice susceptible to the development of autoimmune uveitis [38].
Autoimmune diseases are more frequent in aged and in female individuals [39] and thus we asked whether Hsp65 interference in autoimmunity is age and/or gender-related. Reports of Hsp65 interference in autoimmunity and other biological alterations occurring during the immunosenescence process are related to gender and aging [40]. These findings lead us to investigate whether M. leprae Hsp65 is also involved in alterations of aged individuals, as the immunosenescence process can lead to the onset of autoimmunity. It was assessed the role played by passive administrations of WT and mutant K 409 A Hsp65 on the lifespan and antibody production of aged H III and L III mice. We conclude that the WT protein administration interferes with the survival of aged and adults H III female mice, even though the anti-DNA and anti-Hsp65 antibody production was not markedly changed. As no significant changes in male mice survival and antibodies production were observed we conclude that Hsp65 effects were gender-related.

WT Hsp65 administration reduces the lifespan of H III female mice
Male and female two hundred-seventy-days old (aged) mice were inoculated intraperitonially with a single dose of 2.5 μg/animal of WT or K 409 A Hsp65 in 0.2 mL of PBS, or only PBS as control group. Mice were observed until death for mean survival time (MST) and environmental variance (V E ) determination (Table 1). Figure 1A illustrates the survival reduction of aged H III female mice inoculated with the native protein (308 ± 25 days, p < 0.01) compared to controls (530 ± 123 days). Also it was observed a decrease of 24-fold in the WT group phenotypic variance (V E = 635 days in WT group versus V E = 15293 days in control group). Conversely, Figure 1B shows that aged L III females inoculated with WT Hsp65 presented higher MST (615 ± 46 days; p < 0.01) compared to K 409 A (442 ± 97 days) and control group (441 ± 72 days) and a decrease in V E (2180 days) in contrast with control (5328 days) and K 409 A (9590 days) groups. No survival differences were observed in aged male mice from both H III and L III in all experimental groups ( Figure 1C and D).
Since the lifespan of aged H III females (270-days old) was shortened after WT Hsp65 inoculation, we asked whether the same effect could be observed in adult H III female mice (120-days old) injected with Hsp65 ( Figure 2). Adult H III female mice showed a MST reduction after WT protein administration (466 ± 134 days; p < 0.01) when compared to K 409 A injected group (665 ± 37 days). The first death occurred 247 days after inoculation of the native molecule, and the environmental variance was lower in mutant-(1422 days, 22-fold less) and WT-inoculated animals (18181 days, about 2-fold less) compared to control group (31473 days).
It is noteworthy that in mice from both lines and ages no weight loss, piloerection or ascites were detected.
Anti-DNA and anti-Hsp65 antibody production are altered after WT Hsp65 injection Since the antibody production against heat shock proteins are involved in autoimmune processes, we measured the anti-DNA and anti-Hsp65 IgG1 and IgG2a isotypes production after WT and point-mutated Hsp65 inoculation. In aged H III female mice, WT and K 409 A inoculation caused a 4.5-fold increase in anti-Hsp65 IgG1 production (p < 0.01) ( Figure 3A), compared to pre-immune animals (zero days), at fourteen and seven days post-inoculation, respectively. Despite the reduction observed in IgG2a anti-Hsp65 ( Figure 3A) and IgG1 anti-DNA ( Figure 3B) observed in aged H III female mice, the non-inoculated group (control) also showed this decrease, possibly indicating an environmental interference non-related with the protein injection. No differences in IgG2a anti-DNA production were observed.
Antibody production kinetics analysis of aged H III female mice shows an increase of the IgG1 anti-Hsp65 (p < 0.01) in the WT-injected group, starting at 2.8 log 2 and reaching 5.4 log 2 on day fourteen, and 3.3 log 2 to 6 log 2 in K 409 A group (p < 0.05) on 7 th day post-immunization when compared to pre-immune serum (data not shown).
In adult H III female mice, the administration of both WT and K 409 A Hsp65 molecules did not promote alterations in the anti-Hsp65 antibody production kinetics between all experimental groups until 33 days postinoculation ( Figure 4A). However, the anti-DNA titer was increased at 33 rd days in WT (p < 0.01) and K 409 A (p < 0.05) groups ( Figure 4B). Regarding aged H III male mice, the isotypes titration showed only increased IgG1 anti-Hsp65 isotype production compared to pre-immune groups at the 14 th day after the K 409 A Hsp65 injection (data not shown).
After the observation of the increased survival of aged L III female mice injected with WT Hsp65, we investigated the effects of those molecules over the   antibody production in low antibody responder mice.

Discussion
Immunological functions may change with aging and lead to a deficient immune response to several infections and vaccines, predisposing the individual to chronicdegenerative processes by the decline of self-tolerance maintenance and loss of tissue integrity, directing to crypt antigens release, amplified bystander activation and molecular mimicry events [2,41,42]. Based on the immune alterations observed in the aging process, the reports of higher incidence of chronic-degenerative processes in elders [14,43] and the deleterious effect of M. leprae Hsp65 administration on murine lupus and autoimmune uveitis [35,38], we evaluated the interference of WT Hsp65 administration on survival time and correlation with antibody production in H III and L III mice. The animals from  Selection III are a well-established model to understand the humoral immune response and its influences over the susceptibility to infections [44], autoimmunity [45] and tumorigenesis [46]; they also differ in the response to antigens not related to those used for the selection procedure [47] and present different susceptibility to autoimmune disease [48]. Moreover, H III and L III mice are used to verify the influence of genetic alterations over longevity, as demonstrated by the differences between the survival of distinct genetic bidirectional selections: Selection I and II presents different survival between lines and gender and Selection III shows similar mean survival time regardless of sex or linage (H III : 611 ± 153 days, L III : 622 ± 166 days) [49,50]. Native Hsp65 effects on survival of aged and adults H III female mice indicate the immunological interference of the Hsp65 molecule in this model. It reduced the survival of H III female mice, mainly in young aged (270-days old), with first death occurring 18 days after WT administration, whereas for adults H III female mice (120-days old) it occurred 247 days post-injection, a period 14-times greater than the aged mice. In opposition, WT-inoculated aged L III female mice presented higher MST (615 ± 46 days, p < 0.01) compared to mutant and control groups (442 ± 97 and 441 ± 72 days, respectively). Despite the differences of the maximal survival time in control groups (Additional file 1), we can take these data in account as the maximal survival obtained was within to those previously observed [50], with 501 days in aged L III female and 759 days in aged L III male mice. The maximal survival of control group of H III female (763 and 765 days) or L III male (759 days) showed a significant difference because some mice died with extreme lifespan. Despite the low number of animals used (restricted by the ethics committee) the results demonstrated that the M. leprae Hsp65 injection could alter the survival, reducing MST in H III and augmenting in L III female mice.
Considering that L III line shows an 18-fold reduction in the antibody production and low T cell proliferation [51], it is possible to consider that the humoral response of L III mice against Hsp65 is reduced, resulting in an easy management by the system to return to homeostasis. The same could be true for other mice lines with lower antibody response compared to H III mice, like Swiss albino mice (foundation of Selection III lines) [47]. Contrary, the high responder rate of the immunological system in H III female mice after WT injection, despite the antibody production rate, could disrupt the homeostasis and lead to a reduced survival. Homeostasis imbalance due to aging process in association with the interference of Hsp65 inoculation could explain the higher decrease in survival of H III aged female. In spite of our animal model do not spontaneously develop autoimmune processes, the reduced survival time of H III female mice matches to the experiments involving lupus-prone mice [35] and models of experimental autoimmune uveitis in mice [38], and reassure the involvement of Hsp in chronic-degenerative processes. The control expression and the rupture of Hsp65 balance in SLE development were ascertained through the approach of inductive disequilibrium of physiological and immune states by homologous Hsp [52] and the same could be true for the current study.
Since High and Low mice lines differ, respectively, to high or low antibody responses, the anti-DNA and anti-Hsp65 IgG isotypes production were analyzed after WT or K 409 A Hsp65 inoculation. H III and L III mice, both genders, presented slightly higher production of IgG2a anti-Hsp65 and anti-DNA related to the IgG1 isotype. This balance towards a Th1 response may indicate a natural pro-inflammatory status in these strains which is confirmed by the relative easy way to induce adjuvant arthritis in H III mice [53]. The time-course analyses of immunoglobulin production did not show significant intragroup differences, which might be related to the absence of a strong specific response to these proteins demonstrated by the low antibody titers even 30 days after the protein inoculation and because this is not an immunization process. Compared to pre-immune mice, aged H III females presented an increase of anti-Hsp65 IgG1 in the group treated with WT or K 409 A Hsp65 and a non-related reduction of IgG2a. In murines, the IgG1 and IgG2a functions are dependent on the cells activation threshold determined by the affinity of antibodies and the expression of inhibitors/activators receptors [54]. The amplification of anti-Hsp65 IgG1 antibody (approximately 4-times) should be related with a switch to a Th2 response, previously observed in the immunosenescence process [2]. It also could correlate to the precocious death of aged H III female mice, since the augment of Th2 cytokines, despite its regulatory effect, are involved in some diseases like asthma, allergies and autoimmunities [55,56].
Confirmed the M. leprae Hsp65 effect on reduced survival in aged H III female mice, it was evaluated whether the anti-DNA and anti-Hsp65 antibody production were agedependent by comparing the IgG isotypes with adults H III female mice. They did not present intergroup changes in kinetics of anti-DNA or anti-Hsp65 antibodies production, Figure 5 Anti-Hsp65 and anti-DNA antibodies production in aged L III mice. Anti-Hsp65 IgG1 and IgG2a (A and C) and anti-DNA IgG1 and IgG2a (B and D) isotype production in the serum of aged L III mice (4-8 animals/group, males and females) inoculated with WT, K 409 A Hsp65, or PBS (control). Antibody titers were determined by ELISA before (day 0) and after 7 and 14 days post-WT or -K 409 A Hsp65 inoculation. *p < 0.05, **p < 0.01 and ***p < 0.001 (Two-way ANOVA, Bonferroni multiple comparisons post-test).
but adult H III female mice showed increase in IgG2a titers after Hsp65 molecules injection. The IgG2a isotype titers are remarkably lower in adult mice compared to aged ones as an indicative of a better balance of Th1/Th2 response and maintaining homeostasis of the immune system, as demonstrated by the 13-times later death in adult H III females after WT inoculation compared to old mice.
This dominating effect observed in the survival time of aged mice emphasizes the involvement of the Hsp65 molecule in aging processes. The effect of native molecule was gender-specific, as demonstrated by the unaltered MST in aged male mice (H III and L III ) inoculated with both proteins, and potentially related with the differential regulation of the immunological system by sexual hormones [57], as the dimorphism between gender is positively linked with different susceptibility for infections, autoimmune diseases and tumor incidence [40,58]. Sexual hormones (mostly estrogen but also progesterone and testosterone) affect immune cells quantitatively and qualitatively and impact on cytokine production [59]. Females have higher plasma concentrations of immunoglobulin, increased number and strong activation of CD4 + T cells, elevated levels of Th1 cytokines (IgG2a production) and stronger primary and secondary antibody response [60,61]. Indeed, the higher incidence of SLE in females reflects the gender dimorphism [62,63].
It should also be considered the animal model used to test the relationship of Hsp and aging. The opposite phenotypes of antibody production in H III and L III mice include immune response to a wide range of antigens [47]. The F 0 -foundation population -of Selection III mice are genetically heterogeneous: the phenotypic variance (V P ) is due to the sum of the genetic variance (V G ), and the environmental variance (V E ) emerges by all the causes of variability resulting from the environment. The bi-directional selection resulted in a progressive increase (H III mice) or a decrease (L III mice) in mean antibody response, followed by the decline of V G in both lines [47,64]. Therefore, the alterations provoked by WT or K 409 A Hsp65 administration (the environmental feature applied during the experiments) provide the variance in our experimental model. After WT inoculation, a great reduction in variance value was observed in aged female H III (V E = 635 versus V E = 15293 of the control group) than L III mice (V E = 2180 versus V E controls = 5328), showing the impact of this molecule administration over the potential of death phenotype (MST = 308 ± 25 in aged H III and MST = 615 ± 46 in aged L III female mice). We cannot exclude the interference of others genetic factors occurred during the selective breeding, and a gender-effect that may affect the response in H III and L III mice. Since it was proposed that the presence of anti-Hsp60 autoantibodies, innate risk factor of atherosclerosis in adulthood, may be an inherited trait, we are conducting studies about the effect of Hsp65 in (H III xL III )F 1 hybrids to clarify the genetic influence of this susceptibility [65].
Mechanisms underlying the distinct effects of the native Hsp65 on survival of H III and L III mice, and the comparison between them and mutant injected individuals are under evaluation. Preliminary histopathological analysis with some control-and WT-injected H III female mice (3 animals/ group) indicates that the WT Hsp65 inoculation results in a widespread chronic hepatitis, spleen hyperplasia, and, unlike L III female, higher degree of nephrosis and chronic nephritis with inflammatory infiltration of plasma cells, macrophages and lymphocytes (data not shown), characteristics also present in human lupus nephritis [66]. Studies of immune cells alterations in spleen and blood samples of aged H III female mice injected with M. leprae Hsp65 are in progress. The initial results shows increased splenic B cells percentage, amplified expression of CD45RA and CD154 in CD4 + T cells, reflecting on the augmented anti-Hsp65 IgG1 isotype production observed here, and amplified surface expression of CD11b and CD11c in blood monocytes.
The adaptive management of biological systems according to environmental changes is essential for the organism survival and Hsp molecules can interfere with immune phenotypes submitted to independent polygenic control. The aging process presents an impaired cellular homeostasis and the Hsp presentation by antigen presenting cells may be diminished, being responsible for the decline in immunoregulation through the recognizing of self Hsp [67]. On the other hand, an amplified expression of stress proteins and his antigenic determinants can evolve to a pathogenic or regulation of chronic-degenerative processes [68][69][70]. Taking together, these facts explain the pleiotropic effects of Hsp65 on biological systems and its wide actions over different cell types and production of other molecules. Based on pleiotropy, the Theory of Aging proposed by George C. Williams suggests that some genes responsible for increased fitness in young fertile individuals may contribute to the reduction of such capacity in later life [71]. Then, it is conceivable that selection for high antibody production genes, essential for the immune protection through the life of an individual, can be one of the factors that allows Hsp65 act on the immune or physiological imbalance later in life. As previously reported by our group, the WT Hsp65 passive administration affects the endogenous balance by increase the entropy of the individual system; the linear equation proposed (y = a + Δi) shows that the immunological history (y) is unique, irreversible and cumulative [35]. In this study, the animal model employed is not naturally predisposed to autoimmunity, so "a" should include the potential advent of chronic-degenerative processes in aging and "i" consists by the sum of the environmental factors that modulate the entropy: age, gender, antibody production rate, and possible cellular and molecular alterations established in Selection III after Hsp65 administration. All these elements interfere in how Hsp65 interact with the immune system; consequently, the greater their influence, greater the entropic energy, hindering the recovery of homeostasis and contributing to the deaths of H III female mice.
Despite the absence of strikingly differences in antibody production in our experiments, perhaps the 2-fold higher antibodies production in H III females compared to males [47], associated with the senescent immune system and influenced by hormones, are sufficient to induce frailty after WT administration. In parallel, the high antibodies production rate in H III , besides increasing the system entropy, can result in reduced antibody affinity for the protein, facilitating its subsequent binding to self-antigens. In case of imbalance due to the overstimulation by stress or inflammation, autoimmune diseases may emerge or aggravate [35,52]. The opposed occurs in L III females, which presented increased survival when injected with the native molecule, suggesting that the low immunoglobulin production may favor the control of immune system overstimulation.
We do not observed any signs of disease development during the survival time assay and this be correlated with the incapacity of the mycobacterial Hsp65 alone to induce, in some cases, autoimmunity. In an experiment of arthritis induction by Complete Freund's adjuvant (CFA) replaced with the whole mycobacterium [72], the intradermal injection induced arthritic lesions at the same degree as CFA in ankle joints, with the production of anti-DNA and anti-Hsp65 in rats. Thus, the not remarkable increase of anti-Hsp65 antibodies presented by H III and L III mice may be responsible for the absence of disease.
More than a phenotypic effect by the antibody production against WT Hsp65, the extended pleiotropic effect of this protein over the immune system results in strengthening of naturally established disorders in aged H III female mice who possibly present a natural higher degree of entropy. In addition, the age-remodeled immune system already shows a major entropy level and the injection of M. leprae WT Hsp65 in females reinforce an imbalance that does not resemble the young individuals, originating disorganizations and irreversible processes leading to death.

Conclusions
Here we verified in an aging mice model the role of M. leprae Hsp65 in the aggravation of phenotypes, as observed in SLE and experimental autoimmune uveitis, and outlined its interference mainly in aged H III female inducing precocious death. We assume that this effect is associated to the aging process and related to gendereffect instead of the amount of antibody produced in these mice lines. Studies of cellular and cytokines alterations after the Hsp65 administration in Selection III mice and its (H III xL III )F 1 hybrid mice are in progress to elucidate the mechanisms by which this heat shock protein and its responses act in the immune system of aged individuals.

Methods
Expression of the recombinants M. leprae Hsp65 in Escherichia coli and purification Clone pIL161, containing the coding sequence of the M. leprae WT Hsp65 and its point-mutated form K 409 A [36] were amplified in E. coli DH5a cells. Expression and purification of the recombinant Hsp65 WT and K 409 A was described elsewhere [35].

Animals
The genetically heterogeneous selected mice for High (H III ) or Low (L III ) antibody production were bred in the animal facility of the Immunogenetic Laboratory and maintained at the animal facility of the Immunochemistry Laboratory, both in Butantan Institute. They were housed in groups of four to five in plastic cages filled with hardwood bedding, provided with water and rodent chow ad libitum, in a room with 12-h light/dark cycle, controlled pressure, humidity and temperature (24 ± 2°C). All procedures are in agreement to the International Animal Welfare Recommendations [73] and the experiments are in conformity with the Ethical Principles in Animal Research, adopted by the Brazilian College of Animal Experimentation, and was approved by the Ethical Committee for Animal Research of Butantan Institute (CEUAIB #475/08).

Administration of the WT and K 409 A Hsp65 molecules
Male and female H III and L III mice at the age of 120-or 270-days old (named here "adult" or "aged" mice, respectively) were inoculated intraperitonially with a single dose of 2.5 μg of WT or K 409 A Hsp65 of M. leprae in 0.2 mL of phosphate buffer saline pH 7.4 (PBS); as controls, mice were injected with 0.2 mL PBS. In this study it is important to highlight that the periodically bleedings were performed at different time points in aged and adult female mice; from previously observations that aged H III female animals death started 18 days after the WT Hsp65 administration (bleedings occurred at seven and at fourteen days post proteins administration). Differently, adult H III female individuals, which presented an extended survival time after the WT Hsp65 inoculation, were bled after fourteen and thirty-three days after the proteins injection; this longer interval was used to avoid external stress stimuli that could influence the experiment. The serum samples were stored at −20°C until antibody titration. Each animal was observed until death for the designing of the longevity curve and examined for clinical signs that include development of ascites and lethargy.