Amniotic fluid heat shock protein 70 concentration in histologic chorioamnionitis, term and preterm parturition

Abstract

Objective. Heat shock protein (HSP) 70, a conserved member of the stress protein family, is produced in almost all cell types in response to a wide range of stressful stimuli, and its production has a survival value. Evidence suggests that extracellular HSP70 is involved in the activation of the innate and adaptive immune response. Furthermore, increased mRNA expression of HSP70 has been observed in human fetal membranes following endotoxin stimulation. This study was conducted to determine the changes in amniotic fluid HSP70 concentrations during pregnancy, term and preterm parturition, intra-amniotic infection (IAI), and histologic chorioamnionitis.

Study design. A cross-sectional study was conducted in 376 pregnant women in the following groups: (1) women with a normal pregnancy who were classified into the following categories: (a) women in the mid-trimester (14–18 weeks) who underwent amniocentesis for genetic indications and delivered normal infants at term (n=72); (b) women at term not in labor (n = 23); and (c) those at term in labor (n = 48). (2) Women with spontaneous preterm labor and intact membranes who were subdivided into the following categories: (a) preterm labor who delivered at term without IAI (n = 42); (b) preterm labor who delivered preterm without IAI (n = 57); and (c) preterm labor and delivery with IAI (n = 30). (3) Women with preterm prelabor rupture of membranes (PROM) with (n = 50) and without (n = 54) IAI. Among patients with preterm labor with intact membranes and preterm PROM who delivered within 72 hours of amniocentesis, placenta, umbilical cord, and chorioamniotic membranes were collected and assessed for the presence or absence of acute inflammatory lesions in the extraplacental membranes (histologic chorioamnionitis) and/or umbilical cords (funisitis). HSP70 concentrations in amniotic fluid were determined using a sensitive and specific immunoassay. Non-parametric statistics were used for analysis. A p value of <0.05 was considered statistically significant.

Results. Immunoreactive HSP70 was detected in 88% (332/376) of amniotic fluid samples. The median amniotic fluid HSP70 concentration was significantly higher in women at term without labor than in those in the mid-trimester (term no labor: median 34.9 ng/mL, range 0–78.1 ng/mL vs. mid-trimester; median 6.6 ng/mL, range 0–20.8 ng/mL; p<0.001). Among patients with spontaneous preterm labor and preterm PROM, those with IAI had a significantly higher median amniotic fluid HSP70 concentration than those without IAI (preterm labor with IAI: median 82.9 ng/mL, range 0–500 ng/mL vs. preterm labor without IAI: median 41.7 ng/mL, range 0–244 ng/mL; p = 0.001; preterm PROM with IAI: median 86.5 ng/mL, range 0–428 ng/mL vs. preterm PROM without IAI: median 55.9 ng/mL, range 14.9–299.9 ng/mL; p = 0.007). There was no significant difference in the median amniotic fluid HSP70 concentration between patients with preterm labor who delivered preterm without IAI and those who delivered at term (p = 0.6). However, among patients with preterm labor without IAI, there was an inverse relationship between amniotic fluid concentration of HSP70 and the amniocentesis-to-spontaneous delivery interval (Spearman's Rho = −0.26; p = 0.02). Patients with histologic chorioamnionitis/funisitis had a significantly higher median amniotic fluid HSP70 concentration than those without inflammation (inflammation: median 108.7 ng/mL, range 0–500 ng/mL vs. without inflammation: median 67.9 ng/mL, range 7.1–299.9 ng/mL; p = 0.02). Women at term in labor had a median amniotic fluid concentration of HSP70 significantly higher than those not in labor (term in labor: median 60.7 ng/mL, range 0–359.9 ng/mL vs. term not in labor: median 34.9 ng/mL, range 0–78.1 ng/mL; p = 0.02).

Conclusions. Intra-amniotic infection, histologic chorioamnionitis, and term parturition are associated with elevated amniotic fluid HSP70 concentrations. HSP70 plays a role in the host defense mechanism by activating the innate arm of the immune response in women with intrauterine infection. The mechanisms of preterm and term parturition in humans may involve extracellular HSP70.

Keywords:

• HSP70
• intra-amniotic infection
• preterm labor
• preterm prelabor rupture of membranes
• histologic chorioamnionitis
• parturition

Introduction

Heat shock proteins (HSPs) are highly conserved molecules [1],[2] that are present in almost all sub-cellular structures (e.g., nucleus, mitochondria, endoplasmic reticulum, and cytoplasm) of all cell types from prokaryotes to eukaryotes [2]. HSPs regulate intracellular processes to maintain homeostasis during cell proliferation/differentiation and thus, function as molecular chaperones [3-6]. An increased expression of intracellular HSPs is observed following cell exposure to stressful stimuli such as hypoxia, ischemia, and high temperature [7],[8]. HSPs are categorized into several families according to their approximate molecular weight (e.g., HSP40, HSP60, HSP70, HSP90, and HSP110). Among all HSPs, HSP70 is the best characterized [9].

HSPs also participate in innate and adaptive immune responses, and were originally considered to be exclusively intracellular proteins. Their presence in the extracellular compartment reflects tissue damage or ‘danger signals’[10]. HSPs released from necrotic cells have been proposed to activate monocytes through diverse cell-surface receptors (CD14, Toll-like receptor (TLR), CD40, etc.) [11-17], which in turn stimulate the production of pro-inflammatory cytokines [18-20]. HSPs participate in antigen processing and presentation by antigen presenting cells (APC), which elicit a robust T cell response in adaptive immunity [21]. However, recent evidence suggests that HSPs can be released from cells without necrosis [22],[23]. Indeed, rat glia cells [24], human islet cells [25], and human peripheral blood mononuclear cells [26] have been shown to release HSPs by exocytosis in the absence of detectable cell death [23]. HSP60 [27] and HSP70 [28] are normally present in serum of healthy individuals. However, psychological stressors have been shown to increase circulating HSPs in animal experiments [29,30] and changes in HSP concentrations in blood have been reported in several pathological conditions. Elevation of serum HSP60 concentration was observed in patients with early atherosclerosis [31], and high serum concentrations of HSP70 were reported in patients with peripheral/renal vascular disease [32] and in those with preeclampsia [33].

Infection is an important mechanism of disease in preterm parturition [34-40]. Indeed, it is the only pathologic process for which a firm causal link with prematurity has been established and a defined molecular pathophysiology is known [41]. Moreover, intra-amniotic infection/inflammation has been implicated in the genesis of fetal and neonatal injury [42],[43] leading to cerebral palsy [44] and chronic lung disease [45]. Several lines of evidence suggest a role for HSP70 in preterm labor. Among patients who were at risk for preterm delivery, the mean serum concentration of HSP70 was higher in patients who delivered preterm than in those who delivered at term [33]. Moreover, increased mRNA expression of HSP70 was observed in cultured human amnion following endotoxin stimulation [46]. Finally, HSP70 antigen–antibody complexes were detected in the placenta of some patients who delivered preterm [47]. There is no previous information on HSP70 concentrations in the amniotic cavity of patients with preterm labor.

The aim of this study was to determine the changes of amniotic fluid HSP70 concentrations throughout gestation, during parturition in term and preterm pregnancies, and in the presence of intra-amniotic infection (IAI) and histologic chorioamnionitis.

Materials and methods

Study design and population

A cross-sectional study was conducted by searching our clinical database and bank of biologic samples. A total of 376 women were classified into three groups: (1) women with a normal pregnancy who were further subdivided into the following categories: (a) women in the mid-trimester (14–18 weeks) who underwent amniocentesis for genetic indications and delivered normal infants at term (n = 72); (b) women at term not in labor (n = 23); and (c) women at term in labor (n = 48). (2) Women with spontaneous preterm labor and intact membranes who were subdivided into the following categories: (a) preterm labor who delivered at term without IAI (n = 42); (b) preterm labor who delivered preterm without IAI (n = 57); and (c) preterm labor and delivery with IAI (n = 30). (3) Women with preterm prelabor rupture of membranes (PROM) with (n = 50) and without (n = 54) IAI.

Clinical definitions

Preterm labor was diagnosed by the presence of at least two regular uterine contractions every 10 minutes associated with cervical changes that required admission to the hospital before 37 weeks of gestation. IAI was defined as an amniotic fluid culture that was positive for microorganisms. The results of the amniotic fluid analyses were used for clinical management. Women at term not in labor underwent amniocentesis for the assessment of fetal lung maturity prior to cesarean section. Women at term in labor consisted of women who were suspected to have preterm labor because of uncertain dates and had an amniocentesis for the assessment of microbial invasion and fetal lung maturity. However, if they delivered a neonate greater than 2500 grams without complications of prematurity, they were considered likely to represent patients in spontaneous labor at term. PROM was defined as amniorrhexis before the onset of spontaneous labor. Membrane rupture was diagnosed by vaginal pooling, ferning, or a positive nitrazine test.

Sample collection

Amniotic fluid collection was performed by trans-abdominal amniocentesis under ultrasonographic guidance. Amniotic fluid was transported to the laboratory and cultured for aerobic/anaerobic bacteria as well as genital mycoplasmas. White blood cell (WBC) count, glucose concentration, and Gram stain for microorganisms were performed in amniotic fluid. Among patients with preterm labor with intact membranes and preterm PROM, placenta, umbilical cords, and chorioamniotic membranes were collected. The presence or absence of acute inflammatory lesions in the extraplacental membranes (histologic chorioamnionitis) and/or umbilical cords (funisitis) in those who delivered within 72 hours of amniocentesis was assessed as previously described [48]. This period of time was selected to preserve a meaningful temporal relationship between amniotic fluid HSP70 concentrations and membrane pathologic findings.

All women provided a written informed consent prior to the collection of samples. The Institutional Review Boards of Wayne State University, and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD/NIH/DHHS) approved the collection and utilization of samples for research purposes. Many of these samples have been employed to study the biology of cytokines [49-52], chemokines [53], antimicrobial peptides [54], and growth factors [55] in normal pregnant women and in those with pregnancy complications.

HSP70 immunoassays

The immunoassay kits (Stressgen Biotechnology Corporation, Victoria, BC, Canada) are specific for both native and recombinant HSP70. The HSP70 immunoassay was validated for human amniotic fluid in our laboratory. Briefly, amniotic fluid samples were incubated in duplicate wells, pre-coated with monoclonal antibodies to an inducible form of HSP70. The HSP70 protein were detected by a biotinylated monoclonal antibody. The final step involved signal amplification based on a biotin–avidin coupling in which avidin was linked to horseradish peroxidase. The amount of HSP70 was measured upon addition of tetramethylbenzidine utilizing a programmable spectrophotometer (Ceres 900 Micro plate Workstation, Bio-Tek Instruments, Winooski, VT, USA) set to read absorbance at 450 nm. Amniotic fluid HSP70 concentrations were derived from interpolating the absorbance readings from a standard curve generated from known concentrations of HSP70. The inter- and intra-assay coefficients of variations were 6.7% and 4.4%, respectively. The sensitivity was 0.7 ng/mL.

Statistical analysis

Kruskal–Wallis and Mann–Whitney U-tests were used to determine the differences in the median amniotic fluid HSP70 concentrations among and between groups, respectively. Spearman rank correlation was utilized to assess correlations between amniotic fluid concentrations of HSP70, glucose, and WBC count. A p value of <0.05 was considered statistically significant. Analysis was performed with SPSS software version 12.0 (SPSS Inc., Chicago, IL, USA).

Results

Demographic and clinical characteristics

The median gestational age at amniocentesis in patients with preterm labor and intact membranes without IAI, who delivered preterm, was significantly lower than in those who delivered at term (p <0.001, Table I). There was no significant difference in the median gestational age at amniocentesis between patients who delivered preterm with and without IAI (p = 0.2, Table I). Similarly, there was no significant difference in the median gestational age at amniocentesis between patients with preterm PROM with and without IAI (p = 0.3, Table I). Demographic and clinical characteristics of women in the mid-trimester, women at term not in labor, and women at term in labor are displayed in Table II.

Table I.  Demographic and clinical characteristics of patients with preterm labor and intact membranes (PTL) and patients with preterm prelabor rupture of membranes (PROM).

Characteristics	PTL who delivered at term without IAI (n = 42)	p	PTL who delivered preterm without IAI (n = 57)	p^a	PTL with IAI (n = 30)	p^b	Preterm PROM without IAI (n = 54)	p^c	Preterm PROM with IAI (n = 50)
Maternal age (years)	21 (16–39)	0.2	23 (15–38)	0.5	25 (16–33)	0.3	24 (15–41)	0.1	28 (17–39)
Nulliparous	10 (24%)	0.2	21 (37%)	0.4	14 (47%)	0.04*	18 (33%)	0.9	16 (32%)
Race
African-American	37 (88%)	0.8	49 (86%)	0.7	27 (90%)	0.4	43 (80%)	0.2	46 (92%)
Caucasian	3 (7%)		5 (9%)		2 (7%)		10 (18%)		4 (8%)
Hispanic	2 (5%)		2 (3%)		0		1 (2%)		0
Other	0		1 (2%)		1 (3%)		0		0
Gestational age at amniocentesis (weeks)	30.6 (23.3–34.0)	<0.001*	27.3 (20.6–34.0)	0.2	25.9 (20.0–33.3)	<0.001*	30.5 (20.0–33.7)	0.3	29.9 (20.0–33.9)
Gestational age at delivery (weeks)	39.0 (37.0–41.8)	<0.001*	30.7 (20.9–36.7)	0.002*	26.1 (20.3–34.4)	<0.001*	31.8 (20.0–35.0)	0.003*	29.9 (20.1–34.1)
Amniotic fluid HSP70 detectable	42 (100%)	1.0	56 (98%)	0.3	28 (93%)	0.09	54 (100%)	0.05	46 (92%)

IAI, intra-amniotic infection. Values are expressed as median (range) or number (percent). *p <0.05. ^aComparison between PTL who delivered preterm without IAI and PTL with IAI. ^bComparison between PTL who delivered at term without IAI and PTL with IAI. ^cComparison between preterm PROM with and without IAI.

Table II.  Demographic and clinical characteristics of women at midtrimester, women at term not in labor, and women at term in labor.

Characteristics	Women at midtrimester (n = 72)	Women at term, not in labor (n = 23)	Women at term, in labor (n = 48)
Maternal age (years)	37 (24–42)	27 (17–40)	23 (16–37)
Gestational age at amniocentesis (weeks)	16 (14–18)	39 (38–42)	39 (37–41)
Gestational age at delivery (weeks)	40 (37–42)	39 (38–42)	39 (37–41)
Amniotic fluid HSP70 detectable	41 (57%)	21 (91%)	44 (92%)

Values are expressed as median (range) or numbers (percent).

Changes in amniotic fluid HSP70 concentration during normal pregnancy

Immunoreactive HSP70 was detected in 88% (332/376) of all amniotic fluid samples. However, HSP70 was detected in only 57% (41/72) of women in the mid-trimester. The median amniotic fluid HSP70 concentration was significantly higher in women at term not in labor than in those in the mid-trimester (term no labor: median 34.9 ng/mL, range 0–78.1 ng/mL vs. mid-trimester: median 6.6 ng/mL, range 0–20.8 ng/mL;p <0.001; Figure 1). Women at term in spontaneous labor had a significantly higher median amniotic fluid HSP70 concentration than those not in labor (term in labor: median 60.7 ng/mL, range 0–359.9 ng/mL vs. term not in labor: median 34.9 ng/mL, range 0–78.1 ng/mL; p = 0.02; Figure 2).

Figure 1. Amniotic fluid heat shock protein 70 (HSP70) concentration in women at mid-trimester and at term gestation not in labor. The median amniotic fluid concentration of HSP70 in women at term not in labor was significantly higher than in women at mid-trimester (term not in labor: median 34.9 ng/mL, range 0–78.1 ng/mL vs. mid-trimester: median 6.6 ng/mL, range 0–20.8 ng/mL; p <0.001). LOD: limit of detection. *p <0.05.

Figure 2. Amniotic fluid heat shock protein 70 (HSP70) concentration in women at term gestation. Women in spontaneous labor had a median amniotic fluid HSP70 concentration significantly higher than those not in labor (term in labor: median 60.7 ng/mL, range 0–359.9 ng/mL vs. term not in labor: median 34.9 ng/mL, range 0–78.1 ng/mL; p = 0.02). LOD: limit of detection. *p <0.05.

Changes in amniotic fluid HSP70 concentration during preterm labor and preterm PROM

Patients with IAI and either intact or ruptured membranes had a significantly higher median amniotic fluid HSP70 concentration than those without IAI (preterm labor with IAI: median 82.9 ng/mL, range 0–500 ng/mL vs. preterm labor who delivered preterm without IAI: median 41.7 ng/mL, range 0–244 ng/mL; p = 0.001 and preterm PROM with IAI: median 86.5 ng/mL, range 0–428 ng/mL vs. preterm PROM without IAI: median 55.9 ng/mL, range 14.9–299.9 ng/mL; p = 0.007; Figures 3 and 4). No significant difference in the median amniotic fluid HSP70 concentration was found between patients with preterm labor and sterile amniotic fluid who delivered preterm and those who delivered at term (preterm labor who delivered preterm without IAI: median 41.7 ng/mL, range 0–244 ng/mL vs. preterm labor who delivered at term without IAI: median 38.2 ng/mL, range 7.8–110.7 ng/mL; p = 0.6; Figure 3). Similar results were obtained after adjusting for duration of sample storage and gestational age at amniocentesis using analysis of covariance (p = 0.4). However, among patients with preterm labor who delivered at term and preterm without IAI, there was an inverse relationship between amniotic fluid concentration of HSP70 and the length of the amniocentesis-to-spontaneous delivery interval (Spearman's Rho = −0.26; p = 0.02).

Figure 3. Amniotic fluid heat shock protein 70 (HSP70) concentration in women with preterm labor and intact membranes. The median amniotic fluid HSP70 concentration in woman with preterm labor with intra-amniotic infection (IAI) was significantly higher than in those without IAI who delivered preterm (preterm labor with IAI: median 82.9 ng/mL, range 0–500 ng/mL vs. preterm labor who delivered preterm without IAI: median 41.7 ng/mL, range 0–244 ng/mL; p = 0.001). There was no significant difference in the median amniotic fluid HSP70 concentration between women with preterm labor who delivered preterm without IAI and those who delivered at term (preterm labor who delivered preterm: median 41.7 ng/mL, range 0–244 ng/mL vs. preterm labor who delivered at term: median 38.2 ng/mL, range 7.8–110.7 ng/mL; p = 0.6). LOD: limit of detection. *p <0.05.

Figure 4. Amniotic fluid heat shock protein 70 (HSP70) concentration in women with preterm prelabor rupture of membranes (preterm PROM). The median amniotic fluid concentration of HSP70 was significantly higher in women with preterm PROM with intra-amniotic infection (IAI) than in those without IAI (IAI: median 86.5 ng/mL, range 0–428 ng/mL vs. without IAI: median 55.9 ng/mL, range 14.9–299.9 ng/mL; p = 0.007). LOD: limit of detection. *p <0.05.

Among patients with preterm labor and preterm PROM, there was a positive correlation between the amniotic fluid concentration of HSP70 and WBC count (Spearman's Rho = 0.4; p <0.001), and a negative correlation between the amniotic fluid concentration of HSP70 and glucose (Spearman's Rho = −0.3; p <0.001).

Amniotic fluid HSP70 concentration and histologic chorioamnionitis

Placental pathology was available in 92% (36/39) of patients with spontaneous preterm labor and in 98% (53/54) of those with preterm PROM who delivered within 72 hours of amniocentesis. Patients with evidence of inflammation in the extraplacental membranes (histologic chorioamnionitis) and/or umbilical cords (funisitis) (n = 66) had a significantly higher median amniotic fluid HSP70 concentration than those without inflammation (n = 23) (inflammation: median 108.7 ng/mL, range 0–500 ng/mL vs. without inflammation: median 67.9 ng/mL, range 7.1–299.9 ng/mL; p = 0.02).

Discussion

Principal findings of this study

(1) Immunoreactive HSP70 was present in the amniotic fluid and its concentration increased at term gestation compared to that in the mid-trimester. (2) Patients with IAI (with either intact or ruptured membranes) had a higher median amniotic fluid HSP70 concentration than those without IAI. (3) Amniotic fluid HSP70 concentrations correlated with indirect amniotic fluid markers for intra-amniotic infection/inflammation (WBC count and glucose concentration). (4) Similarly, histologic chorioamnionitis and/or funisitis were associated with higher median amniotic fluid HSP70 concentrations. (5) Women with spontaneous labor at term had a higher median amniotic fluid concentration of HSP70 than those at term not in labor.

Biological activities of HSPs

HSPs were discovered more than 30 years ago and received their names from the observations that the expression of this group of proteins could be induced by high temperatures [56-60]. HSPs are constitutively present in nearly all cell types and considered the most abundant group of molecules in living forms [61]. In humans, there are at least 17 genes encoding for the HSP70 protein family, which are located on various chromosomes [9]. The 73-kDa HSP (HSP73 or heat shock cognate protein 70) is present constitutively, where as the 72-kDa HSP (HSP72 or HSP70) is highly inducible and under the control of the transcriptional factor ‘heat shock factor’[62]. HSP expression is up-regulated by various factors including environmental (heat, ultraviolet radiation [63-69], amino acids [70], heavy metals [71],[72]), physiological (growth factors, cell differentiation, hormonal stimulation) [73-75], pathological (viral, bacterial or parasitic infections [76-78], fever [79], inflammation [80-83], ischemia [84],[85], or autoimmunity [77],[86]) conditions [8],[87],[88].

HSPs function as intracellular molecular chaperones by regulating folding, transportation, translocation, and translation of proteins, which promotes the recovery of cellular activities after stressful stimuli [4],[6],[19],[89],[90]. Moreover, HSPs have an anti-apoptotic activity by inhibiting caspases [91], a group of enzymes that induces programmed-cell death [92]. Evidence in support of this protective function of HSPs includes: (1) HSPs protect human gastric cells from oxidative injury [93], rabbit hearts from ischemic-reperfusion injury [94], and rat retinas from light injury [95]. (2) In an animal experiment, over-expression of HSP72 protects lungs from sepsis-induced injury [96] and is associated with a reduction in hepatocyte apoptosis induced by tumor necrosis factor (TNF)-α[97]. (3) An increased expression of intracellular HSP70 in monocytes and macrophages inhibits TNF-α production following endotoxin stimulation [98]. (4) A preceding heat shock environment, which leads to an increased expression of HSP70, reduces sepsis-induced organ dysfunction and mortality in animal models [99-101]. (5) Polymorphisms of HSP70 gene have been reported in patients with Parkinson's disease, suggesting a protective role of HSP70 against neuronal damage from degenerative disease [102]. Indeed, HSP70 family genes were proposed as candidate genes associated with human longevity [9].

A role of HSP70 in innate and adaptive immunity

HSPs are released extracellularly by either passive or active mechanisms [103]. The passive release results from necrotic cells, while the active release is from viable non-necrotic cells. The release could be as free HSPs, or within exosomes, which are internal vesicles of multivesicular bodies fused with the cell surface [22]. Exosomes can be released as free exosomes or surface membrane-bound HSPs [23]. Subsequently, HSPs bind to specific receptors on the surface of specialized cells including monocytes [104], macrophages [105], B cells [106], dendritic cells [107], and natural killer (NK) cells [103],[108],[109].

Extracellular HSPs can stimulate the innate component of the immune system independently from their chaperone properties. Hence the term ‘chaperokine’ is used to represent the dual roles of HSP [15],[18],[110],[111]. Several studies provide evidence that HSP70 utilizes both TLR-2 (a receptor for Gram-positive bacteria) and TLR-4 (a receptor for Gram-negative bacteria) to induce nuclear factor-kappa B (NFκB) [112] and elicits pro-inflammatory responses in a CD14-dependent manner [12],[19],[20]. HSP70 also participates in antigen processing and presentation by antigen presenting cells [11], resulting in a robust T cell response in adaptive immunity [21]. Soluble as well as membrane-bound HSP70, can directly activate the cytolytic and migratory capacity of NK cells [108],[109],[113].

The human defense mechanisms against many infectious diseases, especially from intracellular pathogens (i.e., Chlamydia trachomatis, Mycobacterium tuberculosis, Plasmodium falciparum), encompass HSP70 as the immunodominant antigen [21]. Immunization with HSPs purified from pathogens has been shown to protect against diseases such as tuberculosis [114],[115], peptic ulcers induced by Helicobacter pylori[116], and infection with Yersinia enterocolitica[117]. Interestingly, the administration of HSP70 purified from tumor cells generates effective anti-tumor specific immunity in animals [118-120].

HSP70 was also proposed to participate in the mechanisms of several autoimmune diseases such as systemic lupus erythematosus [121],[122], rheumatoid arthritis [123],[124], Graves' disease [125], and Hashimoto's thyroiditis [126],[127]. Due to the similarity between eukaryotic and the prokaryotic HSPs, immune recognition of cross-reactive epitopes of pathogens and self-HSPs might be a mechanism linking infections and autoimmune diseases [14]. However, the observations that there are differences in immune responses between pathogens and self-HSPs contradict this view. In an experiment conducted in T-cell lines from synovial fluid of patients with rheumatoid arthritis, T cells stimulated with self-HSP produced Th2-type cytokines (e.g., interleukin (IL)- 4 and IL-10), which were more protective than the Th1-type pro-inflammatory response (e.g., interferon gamma) [128] that was released when T cells were stimulated with bacterial HSP [129].

HSPs in normal pregnancy

The findings that more than half (57%) of normal pregnant women in the mid-trimester and almost all women (91%) at term not in labor had detectable HSP70 in amniotic fluid, support the view that HSP70 can be released extracellularly under physiologic conditions. Our finding is consistent with two previous studies that reported the presence of HSP70 in amniotic fluid during the mid-trimester [130],[131]. Our study also found significantly higher amniotic fluid concentrations of HSP70 in patients at term than in those in the mid-trimester. This phenomenon could be beneficial to pregnant women since HSP70 might function as a ‘chaperokine’ inside the amniotic cavity during growth and development of the fetus. In contrast, there are conflicting reports concerning the changes in HSP70 in maternal serum during pregnancy [33],[132]. While Molvarec et al. [132], in a recent large study, reported a significant increase in serum HSP70 concentration with advancing gestational age, Fukushima et al. [33] did not find significant changes in the mean serum HSP70 concentrations among the three trimesters. The median serum concentration of HSP70 is lower in pregnant than in non-pregnant women [132].

A role of HSP70 in spontaneous labor at term

Our finding that the median amniotic fluid concentration of HSP70 is increased in women with spontaneous labor at term is novel and consistent with a previous observation that HSP70 mRNA expression in sheep myometrium was increased during spontaneous labor [133]. It is noteworthy that the increase in the median amniotic fluid concentration of HSP70 in women at term in labor is modest when compared to the increase observed in women with preterm labor with IAI (60.7 and 82.9 ng/mL, respectively). The proposed mechanism that links an elevation of HSP70 mRNA expression in myometrium and spontaneous labor [133] is that the intracellular HSP70, by binding to the progesterone receptor, functions as a co-repressor of this receptor and suppresses progesterone binding to the nuclear response element [134-136]. However, the precise mechanism leading to an increased HSP70 concentration in amniotic fluid (which is an extracellular compartment) in spontaneous labor at term remains unknown. It is possible that the release of HSP70 from the intracellular compartment might be related to a mild inflammatory response and tissue remodeling process that is frequently observed in the reproductive tract during parturition at term [137-141]. Alternatively, extracellular HSP70 could directly stimulate prostaglandin (PG) production leading to delivery, since HSP70 has been shown to induce cyclooxygenase enzyme (COX)-2 protein expression and PGE₂ production in human umbilical vein endothelial cells [142]. However, there was no information regarding whether HSP70 could stimulate PG production in human amnion.

A role of HSP70 in IAI in preterm labor and preterm PROM

The major finding of this study is that IAI is associated with a higher median amniotic fluid concentration of HSP70. This could be interpreted as reflecting a ‘danger signal’[143],[144] or that HSP70 was released into the amniotic cavity following microbial invasion. This is consistent with the report of Jean-Pierre et al. [131], in which the recovery of Mycoplasma hominis from mid-trimester amniotic fluid was associated with an elevated median intra-amniotic HSP70 concentration. Bacterial endotoxins and HSP70 could engage with TLR-2 and TLR-4 to activate NFκB, and induce the production of pro-inflammatory cytokines including IL-1, IL-6, and TNF-α by mononuclear cells and macrophages leading to PG production and preterm delivery [15],[18],[110],[111]. The relationship between the WBC count and the concentration of HSP70 in amniotic fluid supports this view.

What is the origin of HSP70 in amniotic fluid?

HSP70 protein and mRNA expression has been identified in the epithelium cells in large airways of fetal sheep [145], villous trophoblast, decidua, as well as human chorion and amnion [146]. Menon et al. [46] demonstrated an increased HSP70 mRNA expression in cultured human chorioamniotic membranes after adding endotoxin. These observations suggest that HSP70 could be stimulated and released from chorioamniotic membranes following IAI. Our finding that patients with evidence of inflammation in the extraplacental membranes (histologic chorioamnionitis) and/or umbilical cords (funisitis) had a higher median amniotic fluid HSP70 concentration than those without inflammation supports this hypothesis. Similarly, Fukushima et al. [33] reported that the mean serum concentration of HSP70 was higher in patients who delivered preterm than in those who delivered at term. However, there was no information regarding how many of these patients had intra-amniotic infection. In contrast, Divers et al. [147] could not find any changes in protein expression of HSP70, HSP60, and HSP90 in trophoblasts of the basal plate and decidua of women with preterm delivery compared to those with term delivery. Similarly, a study [47] conducted in placentae from 12 women with preterm and 10 with term birth found no difference in protein expression of HSP70, HSP60, and HSP90 in all specimens. Thus, it is likely that the amnion, not the placenta, is the main source of an increased HSP70 concentration in the amniotic fluid of patients with IAI who delivered preterm.

Interestingly, Ziegert et al. proposed that HSP70 antibody might be involved in the mechanism of preterm labor [47]. In their study, HSP70 antigen–antibody complexes were localized in four of the 12 preterm placentae, but in none of the term placentae [47]. Moreover, maternal serum anti-HSP70 immunoglobulin G (IgG) was present in four cases of preterm birth and in no women at term in labor [47]. Indeed, there was a relationship between the concentration of HSP70 IgG and TNF-α, interferon gamma, and secretory leukocyte protease inhibitor in mid-trimester amniotic fluid suggesting that antibodies to HSP70 might modulate inflammatory responses inside the amniotic fluid cavity [131]. Moreover, a case–control study reported a higher median serum concentration of HSP70 antibody at 16 weeks of gestation in mothers whose neonates were subsequently born with birth defects (cleft lip, cleft palate, and neural tube defects) than that in mothers who gave birth to healthy neonates [148]. Collectively, the roles of HSP70 antibody or antigen–antibody complex in preterm labor require further investigation.

Conclusion

In summary, we report herein that intra-amniotic infection, histologic chorioamnionitis, and term parturition are associated with increased amniotic fluid HSP70 concentrations. HSP70 plays a role in the host defense mechanism by activating the innate arm of the immune response in women with intrauterine infection and may participate in the mechanisms of preterm and term parturition.

Acknowledgement

This research was supported in part by the Intramural Research Program of the Eunice Kennedy Shriver National Institute of Child Health and Human Development, NIH, DHHS.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.