Inherited anxiety-related parent–infant dyads alter LHPA activity

Abstract

The pathogenesis of post-traumatic stress disorder (PTSD) is incompletely understood. We hypothesize that disruptions in mother-child relations may be a key contributor to development of PTSD. A normal and healthy separation-individuation process requires adaptations of self- and interactive contingency in both the mother and her child, especially in early childhood development. Anxious mothers are prone to overprotection, which may hinder the individuation process in their children. We examined long-term stress hormones and other stress markers in subjects three generations removed from the Holocaust, to assess the long-term consequences of inherited behavioral and physiological responses to prior stress and trauma. Jewish subjects who recalled overprotective parental behavior had higher hairsteroid-concentrations and dampened limbic-hypothalamic-pituitary-adrenal (LHPA) axis reactivity compared to German and Russian-German subjects with overprotective parents. We suggest that altered LHPA axis activity in maternally overprotected Jewish subjects may indicate a transmitted pathomechanism of “frustrated individuation” resulting from cross-generational anti-Semitic experiences. Thus measurements of hairsteroid-concentrations and parenting practices may have clinical value for diagnosis of PTSD. We propose that this apparent inherited adaptivity of LHPA axis activity could promote higher individual stress resistance, albeit with risk of an allostatic overload.

Keywords:

• PTSD
• individuation
• Holocaust
• hairsteroids
• cortisol
• Jews

Introduction

Acute stress may lead to anxiety disorders, and particularly to post-traumatic stress disorder (PTSD) with symptoms of dissociation, intrusion, mood changes as well as alterations in arousal and reactivity (APA, 2013). The psychopathology of PTSD may originate from childhood maltreatment (Jaffee, 2017). Biological contributors to PTSD (i.e. “nature”), including structural, hormonal, and genetic changes have been explored extensively (Ullmann et al., 2018), whereas the possible developmental contributors to PTSD (i.e. “nurture”) such as emotional abnormalities, particularly in childhood, have thus far received little attention.

Parental rearing behavior influences an individual’s development across the entire lifespan (Schaefer, 1965). Child-rearing practices, attitudes, and goals are classified as authoritative, permissive, or authoritarian parenting styles (Baumrind, 1967). The most prominent models of child-rearing found in the literature focus on two or even three dimensions. Two factor-analytically derived dimensions are characterized by conceptually opposing pairs independent of each other (Schaefer, 1965; Schumacher, Stobel-Richter, Strauss, & Brahler, 2004). The first pair is labeled as affection (warmth, love) versus rejection (hostility), and the second pair is control (overprotection) versus autonomy (individuation; Schumacher, et al., 2004). Some child-rearing styles are described as factors arising from an abnormal condition or a disease, especially depression and anxiety, but eating disorders as well (Ihle, Jahnke, Heerwagen, & Neuperdt, 2005; Perris, Jacobsson, Lindstrom, von Knorring, & Perris, 1980; Petrowski, Brahler, & Zenger, 2014). Psychological control can stifle independence by restricting the space a child needs to explore and express his or her individuality. Dependency-oriented psychological control creates a dysfunctional dependence. When parents use intrusive and manipulative tactics to maintain the physical and emotional proximity to their children, the overprotective/over-controlling adult often reports a strong fear of abandonment and loss of importance as the child progresses through the individuation process (Kins, Soenens, & Beyers, 2012).

Contemporary developmental theories see the emerging sense of self as an essential organizing function of early development. Stern (1985) described phases of self-perception (sense of emerging self, core self, subjective self, and verbal self) during infancy and toddlerhood (Stern, 1985), which are interwoven with a growing differentiation of relatedness and inter-subjectivity (Trevarthen & Aitken, 2001). In this context, Beebe, Lachmann, and Jaffee (1997), have shown by their micro-analytic observations that parent-infant interactions, which shape the early organization and expectations of experience, are characterized by the salience of arousal, affect, space, and time (Beebe, Lachmann, & Jaffee, 1997). This interactional process is not linear but displays a sequence of state-transforming, facial mirroring, disruption, and repair. The toleration of disruptions and the subsequent repair processes, in particular, provide the infant with experiences essential for the individuation process (Mahler, 1963). An advanced model of impaired mother-child dyads (relationships) expands the concept of self- and interactive regulation to self- and interactive contingency. In this scenario, expectations are generated individually to the extent that the rhythm of one’s behavior can be anticipated from moment to moment as well as separately for each partner (Beebe, 2014).

Parental behavior influences LHPA axis activity, and animal studies have demonstrated that prompt maternal care leads to dampened LHPA axis activity in pups (Liu et al., 1997). Maternal and infant cortisol levels influence one another bi-directionally, especially in depressive mother–child dyads without positive parenting (Hendrix, Stowe, Newport, & Brennan, 2017). A mother's cortisol reactivity following marital conflict could predict her infant's cortisol reactivity, thus providing evidence of mother-to-infant adrenocortical transmission (Hibel & Mercado, 2017). However, there are inconsistent results regarding LHPA axis responses to inhibited individuation arising from overprotective parenting (Narita et al., 2012; Ullmann et al., 2017), and it remains unclear how the LHPA axis responds if maternal care prevents infantile individuation.

Much research focuses on the mechanisms by which stressful and traumatic experiences are transmitted across generations. Descendants of Holocaust survivors show decreased expression of glucocorticoid receptors after gene methylation, which may explain the higher levels of diabetes and anxiety in Jewish individuals (Korenblum et al., 2005; Yehuda et al., 2016). Individuals in the third generation following the Holocaust having experienced higher levels of overprotective parenting behavior with concomitant increased rates of psychosomatic symptoms (Ullmann et al., 2013). On the other hand, lower rates of war-related PTSD were found in children of fathers who survived the Holocaust (Zerach & Solomon, 2016).

We hypothesized that parental overprotection in child rearing would be positively associated with LHPA-axis reactivity in the general population, as well as in immigrants with a similar migration history as assayed by steroid hormone levels in hair samples. Moreover, we hypothesized a negative association between parental overprotection and LHPA-axis reactivity in the context of an allostatic overload in our high risk pool of Jewish subjects three generations removed from the Holocaust and who have to deal with life-time anti-Semitism.

This subject pool permitted us to study multi-generational effects of inherited behavioral and physiological responses.

Subjects and methods

Sample, recruitment and procedure

In this study, we measured physiological and subjective markers of mental strain and recalled parental rearing behavior (Table 1). The "Jewish" (J) group consisted of 40 members (mean age=25.1 years, range=18–36 years) of a seminar by the German Central Jewish Council in September 2014. These subjects were seminar participants from Jewish congregations, who were born in the former Soviet Union (Ashkenazi Jews), and had immigrated to Germany as political refugees ("Kontingentflüchtling") due to the "national" and "popular" anti-Semitism that existed in the former Soviet Union until 1989 (Messmer, 1997). The "Russian-German" (RG) group included 43 subjects (mean age=25.8 years, range=18–35years) who had emigrated from Russia, recruited from a medical care center. These subjects were also born in the former Soviet Union, and their whole families immigrated since 1989 to Germany after resettling during WWII. The "German" (G) group consisted of 40 German students (mean age=24.1 years, range=18–35 years). Overall we included healthy subjects, whose parents were born after 8 May 1945 and grandparents born before this date, which made our subjects third generation post-WWII. Subjects whose family members were survivors of a concentration camp were not classified in our study because we wanted to avoid re-traumatization in the context of the survey. From all groups, we excluded subjects with known endocrine disorders such as hypo-/hyperthyroidism, Cushing’s disease, and Addison’s disease. All subjects were over 18 years of age, and provided signed informed consent. Our study was conducted in accordance with the guidelines approved by the International Review Board of the medical faculty Carl Gustav Carus, Dresden.

Main outcome measures

Recalled parental rearing behavior

The Recalled Parental Rearing Behavior Questionnaire, in German ("Fragebogen zum erinnerten elterlichen Erziehungsverhalten"; FEE), is a psychometrically validated questionnaire that is used internationally in its English companion version derived from the Swedish "Egna Minnen Beträffande Uppfostran" questionnaire ("My memories of upbringing"; EMBU; Perris et al., 1980; Petrowski et al., 2009). The EMBU and their versions assess adults' memories of their parental upbringing by using 24 questions, each answered separately for both mother and father, on a Likert-type scale with the categories 1 (no, never), 2 (yes, sometimes), 3 (yes, often), and 4 (yes, always). Subjects' answers were used to generate three highly interrelated scales of recalled parental rearing behavior for each parent: “rejection and punishment”, “emotional warmth”, and “control and overprotection” (Petrowski et al., 2009). For our study, we only used the "control and overprotection" scale. This scale assesses parental behavior, which the child perceived as overly thoughtful, blaming, interfering, and constricting, and which reflects a distinct orientation toward effort, performance, and high expectations by the respective parent. The EMBU and FEE have previously been used in studies examining perceived parental rearing behavior in siblings and clinical samples in respect to attachment and relationship characteristics, with good psychometric properties in international usage (Cronbach’s Alpha>0.72; Penelo, Viladrich, & Domenech, 2012; Perris, et al., 1980; Petrowski, et al., 2009).

Anti-semitism measurements

Subjects' mental burden of experiences with anti-Semitism in their country of birth (Soviet Union) and their country of immigration (Germany) was assessed for each subject by using the following individual non-standardized questions: “In my country of birth there is anti-Semitism”; “Also in Germany I personally experienced anti-Semitism”. The questions had to be answered in a Likert-type scale with the categories 1 (yes, full agreement), 2 (rather yes, sometimes), 3 (rather no) and 4 (no agreement), range 1–4.

Hormonal data

The long-term concentrations of cortisol, cortisone, dehydroepiandrostendione (DHEA), progesterone, and testosterone were determined following our published protocol with non-pulverized hair (Gao, Kirschbaum, Grass, & Stalder, 2016). By standardized procedure, hair was sampled from the vertex posterior region at the back of the head, which exhibits the most uniform hair growth rates, cut as close as possible to the scalp using fine scissors (Harkey, 1993). Hair samples were stored in a dry and dark environment at room temperature until analysis. 7.5 mg of hair was used for analysis by LC-MS/MS, except four hair samples were 3.0–7.5 mg and one hair sample was 1.1 mg from the Jewish subjects, and one hair sample from a Russian German subject was 5.2 mg.

Control variables measures

Stress perception

The perceived stress questionnaire (PSQ) is a psychometric self-report questionnaire that assesses subjective stress perception and further processing of stressors using 20 scaled responses (range 1–4) such as: “you find yourself in situations of conflict” or “your problems seem to be piling up”. Dominance and external stressors experienced during the previous four weeks can also be quantified by this test. The PSQ shows good psychometric properties in international usage (Cronbach's Alpha>0.85; Fliege et al., 2005).

Psychosomatic symptoms

The modified Giessen complaint list (GBB) is a psychometrically valid (Cronbach’s Alpha=0.93) self-report questionnaire assessing acute episodes of cardiac-related symptoms, fatigue, and epigastric and back pain, and its aggregate value makes it possible to determine the overall burden of physical complaints (Brahler, Schumacher, & Brahler, 2000).

Anxiety and depression

The 14-item hospital anxiety and depression scale (HADS) is an assessment of depressive and anxiety related symptoms commonly used in clinical practice. Higher scores indicate more anxiety and depressiveness, with cutoff values which are regarded as indicating significant depression and anxiety. The psychometric data on HADS demonstrates the quality of this scale on the national as well as international levels (Cronbach’s Alpha>0.80; Roberts, Fletcher, & Merrick, 2014).

Sense of coherence

The sense of coherence (SOC) test assesses one's ability to deal with stressors, and its abbreviated form, consisting of 9 psychometrically validated items (SOC-9L), shows good psychometric properties (Cronbach's Alpha>0.87; Schumacher, Wilz, Gunzelmann, & Brahler, 2000).

Resilience

The resilience scale is a 13 item (RS-13) psychometrically validated, self-report inventory that assesses containment, independence, and persistence, as well as adaptability, tolerance, and flexibility in two dimensions. We used the abbreviated German version which is widely accepted (Cronbachs Alpha>0.90; Leppert, Koch, Brähler, & Strausz, 2008).

Information concerning sports frequency per week, sports intensity, physical activity, weight and height, alcohol consumption per week, income, age, frequency of hair washing, hair tint-coloring/week, and smoking status, were assessed for each subject.

Statistical analyses

Primary, secondary, and descriptive analyses

Our primary objective stress analyses were the steroid profiles of the third generation post-Holocaust Jewish (N = 40), German (N = 40), and Russian German (N = 43) subjects. Moreover, we measured and compared the concentrations of five related LHPA axis steroid-hormones: cortisol, cortisone, DHEA, progesterone, and testosterone using multivariate analysis of variance (MANOVA). To demonstrate group relations we used Scheffé’s post hoc procedures. Subjective stress indicators (“maternal” and “paternal” overprotection as assessed by FEE) in the German and Jewish subjects were compared by two-sided t-test. Before testing our hypotheses we investigated the group differences for variables influencing LHPA axis activity. In this regard, descriptive analyses and group comparisons (MANOVA) were run on all samples for the psychometrically validated questionnaires (HADS, GBB-24, PSQ, SOC-9L, and RS-13) and hair- and stress-related control variables (BMI = weight/height², age, alcohol consumption/week, hair-coloring and tint, sports- and physical activity, sports frequency/week). Except for alcohol consumption, no significant differences were found in our subjective stress-markers, including diagnosis classified according to ICD-10/DSM 5.

Third analyses

Using the z-transformed main values of cortisol, cortisone, progesterone, testosterone, and DHEA, we statistically generated an aggregated variable labeled “Biomarker.” Using the mean-values of the highly internally correlated (r = 0.793, p= <.001) and validated scores for maternal and paternal overprotective parenting (FEE), we constructed another aggregated variable labeled: “Parental Overprotecting”. The process of aggregation of our highly correlated measurements should lead to an increase in validity. For example, using this method, a mediating role of resilience between the parental rearing behavior (FEE), and psychological symptoms including anxiety and depression, can be confirmed (Petrowski et al., 2014). We used an analysis of variance (ANOVA) for analyzing the group differences of the biomarker aggregate (Figure 1).

Figure 1. Cross-cultural steroid-profiling in hair; Legend: MANOVA for hairsteroids, 43 testosterone- and 29 progesterone-levels below the detection limit (<0.01 pg/mg). Cortisol 6.3 ± 0.7 SE pg/mg (J), 4.3 ± 0.5 SE pg/mg (RG), 4.0 ± 0.4 SE pg/mg (G), [F(2,123 = 5.7, p=.004; J > RG (p=.03), J > G (p=.01)]; cortisone 16.7 ± 1.8 SE pg/mg (J), 14.0 ± 1.1 SE pg/mg (RG), 13.7 ± 1.2 SE pg/mg (G), [F(2,123) = 1.4, p = 0.25]; testosterone 0.8 ± 0.1 SE pg/mg (J), 0.7 ± 0.1 SE pg/mg (RG), 0.6 ± 0.1 SE pg/mg (G), [F(2,123) = 0.80, p = 0.45]; progesterone 1.4 ± 0.3 SE pg/mg (J), 0.7 ± 0.1 SE pg/mg (RG), 2.0 ± 1.0 SE pg/mg (G) [F(2,123) = 1.2, p = 0.29]; DHEA 2.2 ± 0.6 SE pg/mg (J), 1.3 ± 0.3 SE pg/mg (RG), 0.6 ± 0.1 SE pg/mg (G), [F(2,123) = 4.5, p=.01; J > G (p=.013)]. ANOVA for “biomarker” after z-transformation (mean: cortisol/cortisone/testosterone/progesterone/DHEA):0.34 ± 0.18 SE pg/mg (J), −0.14 ± 0.13 SE pg/mg (RG), −0.22 ± 0.14 SE pg/mg (G) [F(2,122) = 4.26, p=.02].

Correlation and regression analyses

Pearson’s correlations were used for quantitative examination of overprotective parental practices and hair-steroid differential associations. We calculated a hierarchical regression analysis predicting hair cortisol concentration by group (Jewish and non-Jewish), “Parental Overprotecting,” and their interaction, controlling for age and gender. Furthermore, we examined experiences of anti-Semitism associated to maternal and paternal child rearing practices of overprotection by using Pearson’s correlation coefficient in Jewish subjects three generations removed from the Holocaust.

Results

Increased stress-markers in Jewish refugees

We analyzed LHPA axis activity of five cholesterine derivates in 3 month hair concentrations from Jewish (J, N = 40), Russian German (RG, N = 43), and German (G, N = 40) subjects using MANOVA and ANOVA (Figure 1). The results showed increased LHPA axis activity in the third generation post-Holocaust Jewish subjects versus German and Russian-German subjects, as follows:

Cortisol: J = 6.3 ± 0.7 SEM pg/mg; RG=4.3 ± 0.5 SEM pg/mg; G = 4.0±0.4 SEM pg/mg

Significance: [F(2,123=5.7, p=.004; J > RG (p=.03), J > G (p=.01)]

DHEA: J = 2.2 ± 0.6 SEM pg/mg; RG=1.3 ± 0.3 SEM pg/mg; G = 0.6 ± 0.1 SEM pg/mg

Significance: [F(2,123) = 4.5, p=.01; J > 2G (p=.013)]

Biomarker: J = 0.34 ± 0.18 SEM pg/mg; RG= −0.14 ± 0.13 SEM pg/mg; G= −0.22 ± 0.14 SEM pg/mg

Significance: [F(2,122) = 4.26, p=.02].

Higher values of parental overprotecting

We calculated “Parental Overprotecting” by using the standardized questionnaire for recalled maternal and paternal overprotective behavior (FEE), and found higher values of maternal overprotection in Jewish subjects (mean 17.8 ± 4.6 SD, N = 37) compared to German subjects (mean=15.8 ± 5.2 SD, N = 40; t = 2.10, p=.04), and higher overall “Parental Overprotecting” (aggregated variable) in Jewish subjects (mean=16.9 ± 4.2 SD, N = 38) compared to German subjects (mean=15.0 ± 4.5 SD, N = 40, t = 1.01, p=.06) without reaching the level of significance.

Influence of parental overprotection on the LHPA axis

Since we found significant differences in hair-steroid concentrations and in parental overprotection, we determined the associations between our aggregated main-variables “Biomarker” and “Parental Overprotecting” using Pearson’s correlations in each of the three groups (Figure 2). This indicates a positive association between parental overprotection and the LHPA axis in the group of German (r = 0.393, p=.01, N = 40) as well as Russian German subjects (r = 0.345, p=.02, N = 43) and a negative association in the group of Jewish subjects (r= −0.374, p=.02, N = 38).

Figure 2. (a) Associations between parental overprotecting and LHPA axis activity/line of regression for aggregated variables “Biomarker” associated to “Parental Overprotecting” in Jewish subjects. (b) Associations between parental overprotecting and LHPA axis activity/line of regression for aggregated variables “Biomarker” associated to “Parental Overprotecting” in Russian German subjects. (c) Associations between parental overprotecting and LHPA axis activity/line of regression for aggregated variables “Biomarker” associated to “Parental Overprotecting” in native German subjects.

A hierarchical regression analysis predicting hair cortisol concentration from group (Jewish and non-Jewish), “Parental Overprotecting” and group × “Parental overprotecting” interaction, controlling for age and gender yielded a significant interaction (p<.001; Table 2).

Table 1. sociodemografic as well as stress- and hair-related variabilities of German natives, Jewish- and Russian-German-immigrants.

		Jewish (J) immigrants (N = 40)	Russian-German (RG) immigrants (N = 43)	German (G) natives (N = 40)	F
Age	Mean ± SD	25.1 ± 4.0	25.8 ± 4.6	24.1 ± 4.4	1.62
	Range	18–36	18–35	18–35
Sex	Male	20	21 (48.8%)	18 (45%)
	Female	20	22 (51.2%)	22 (55%)
Sport frequency	Rates per week	2.2 ± 1.5	2.2 ± 1.2	1.6 ± 1.3	2.19
Sport intensity	Intensity per sport unit	5.4 ± 2.3	4.8 ± 2.3	4.7 ± 2.3	1.26
Physical activity	Whole physical activity	5.7 ± 2.1	5.8 ± 2.1	5.5 ± 2.1	0.19
Hair coloring	Rates per week	1.2 ± 0.4	1.2 ± 0.4	1.2 ± 0.4	0.32
Hair tint	Rates per week	1.1 ± 0.3	1.2 ± 0.4	1.1 ± 0.2	2.18
Alcohol	Drinking age/week	4.1 ± 1.0	4.3 ± 1.1	3.1 ± 0.8	16.91*
BMI	Weight/height²	23.9 ± 3.2	24.1 ± 4.9	22.5 ± 4.9	0.58
HADS	Anxiety/depressiveness	10.6 ± 5.1	12.4 ± 5.7	9.8 ± 5.2	2.65
GBB-24	Physical complaints	12.1 ± 8.4	15.9 ± 12.1	11.2 ± 7.3	2.82
RS-13	Resilience	71.0 ± 9.9	68.0 ± 10.8	71.0 ± 8.3	1.27
SOC9-L	Sense of coherence	45.4 ± 4.6	44.2 ± 6.8	45.3 ± 6.5	0.48
PSQ	Perceived stress	0.1 ± 0.0	0.1 ± 0.0	0.1 ± 0.0	0.14

1 GBB-24 missed in J; 1 RS-13 missed in RG; 1 SOC-9L missed in J; 2 PSQ missed in J; 1 PSQ missed in G; sport frequency missed: 9 in J, 12 in RG and 2 in G; sport intensity missed: 1 in J, 2 in RG, 1 in G; 1 physical activity missed in RG; 1 hair coloring missed in J; 2 hair tint missed in RG; BMI missed: 3 in J, 8 in RG, 30 in G.

*The mean difference is significant at the 0.05 level, data are presented in mean ± SD.

Table 2. Hierarchical regression analysis predicting cortisol concentration from group (Jewish and non-Jewish), parental overprotecting and their interaction, controlling for age and gender.

Steps	Predictor variables	Β	R²	ΔR²
Step 1	Age	0.163	0.038	0.038
	Gender (m = 0/f = 1)	−0.082
Step 2	Group (0=Jew/ 1=Non-Jew)	−0.309***	0.132**	0.094**
	Parental Overprotecting	−0.040
Step 3	Group X Parental Overprotecting	−0.255*	0.177***	0.045*

* p≤.05; **p≤.01; ***p≤.001.

We also assessed anti-Semitic experiences in Germany and the former Soviet Union and their association to maternal and paternal overprotection (Table 3), which indicated a significance association between overprotective maternal parenting and anti-Semitism experiences in the former Soviet Union (r = 0.345, p=.04, N = 37) while the other associations were not significant.

Table 3. Anti-semitism and overprotection by Jewish parents.

		Overprotection by Mother	Overprotection by Father
Experiences of anti-Semitism in former Soviet-Union	Pearson's r	0.345	0.171
	Significance	0.04	0.32
	N	37	36
Experiences of anti-Semitism in Germany	Pearson's r	0.087	0.162
	Significance	0.61	0.35
	N	37	36

3 FEE-overprotective dimensions missed for mother and 4 missed for father.

Discussion

Herein we demonstrate increased steroid concentrations and inverted (dampened) LHPA axis reactivity in Jewish refugees who recalled having received a high degree of overprotective parenting, compared to German natives and Russian-German immigrants. We found that higher parental overprotection co-occurs with increasing anti-Semitic experiences in the country of origin. This relationship suggests a possible pathomechanism underlying anxious mother-child relationships as being associated with overprotective maternal rearing practices. The corresponding psychopathological consequences on the LHPA axis activity in children can lead to externalizing symptoms and various mental disorders, including PTSD and depression (Jaffee, 2017; Ullmann et al., 2018; White et al., 2017).

For this study of the long-term consequences of inherited behavioral and physiological responses to prior stress and trauma, we only included subjects that were born three generations removed from the Holocaust. Over generations, Jewish people have had to bear threats of destruction and persecution from their respective social environments. So, the matrix of Jewish mother–infant dyads – as represented in behavioral practices – is of particular interest for determining the consequences of mental strain and stress as influenced by environmental conditions. Individuals with maternal Jewish background have greater prevalence of diabetes and anxiety at an older age (Korenblum et al., 2005). Yet, there are inconsistent findings concerning trauma-related psychophysiological changes in second and third generation individuals after the Holocaust. Some previous studies found no mental disorders, whereas others found increased rates of depression and psychosomatic symptoms as well as intergenerational gene-methylation within the glucocorticoid receptor (Sagi-Schwartz, van, & Bakermans-Kranenburg, 2008; Ullmann, et al., 2013; Yehuda et al., 2000; Yehuda et al., 2016; Zerach & Solomon, 2016). Still others did not allocate their Jewish cohorts to generations following a burdening event such as the Holocaust, or they used conservative methods to measure steroids, or suffered from inaccurate affiliations to the primary group. We found higher long-term hair-steroid profiles independent of circadian and ultradian rhythmicity in Jewish subjects from the former Soviet Union, compared to Russian-German immigrants and German natives. Circadian and ultradian rhythmicity make it difficult to evaluate the long-term activity state of the LHPA axis, and thus it is highly desirable to have a biomarker comparable to the glycated hemoglobin (HbA1c) value in diabetes mellitus (Bolinder, Ungerstedt, & Arner, 1993). For this purpose, hair steroid profiling offers an innovative and promising approach that is of potential clinical diagnostic value for mental disorders.

While hormonal changes in stress-related disorders have been well researched, the clinical consequences of overprotective parenting across the lifespan remain unclear (Kins et al., 2012). The underlying pathomechanisms of early mother-child dyads were focused on self- and interactive contingency (Beebe, 2014; Hendrix et al., 2017; Hibel & Mercado, 2017), whereas concerning the Jewish population, descriptions of psychoanalytical treatments for descendants of Holocaust survivors described the inhibited individuation in childhood without any corresponding physiological evidence (Fonagy, 1999; Kogan, 2003; Levine, 1982). However, links between symmetrical infant preoccupation with facial coordination at the expense of facial self-contingency, and greater difficulty with the individuation of more dependent mother-child dyads, have been suggested (Beebe et al., 2007). More recent results demonstrate that maternal and infant cortisol levels influence one another bi-directionally, particularly in depressive mother-child dyads without positive parenting (Hendrix et al., 2017). Moreover, an adrenocortical transmission capable of bi-directionally regulating mother–infant cortisol levels appears to be induced by a mother’s response to recent marital conflict (Hibel & Mercado, 2017). The interactional process of state transforming, facial mirroring, and moreover, the toleration of disruptions with subsequent repair processes plays a key role in emotional development during the first year of life parallel to motor development while the toddler distances himself from the mother (Mahler, 1963). The paternal role in the triangulation process may provide a containment object for the mother to support the individuation of the child, and this might be indicated among third-generation post-Holocaust father-child dyads with low-level PTSD symptoms in the offspring (Abelin, 1975; Zerach & Solomon, 2016). However, we did not measure the quality of attachment and/or individuation in our Jewish background parent-infant dyads, and our data should be interpreted with that caveat in mind.

Also, the clinical consequences of overprotective parenting across the lifespan remain unclear. Japanese students who recalled overprotective child rearing styles exhibited reduced hippocampal gray matter volume and dampened LHPA axis activity, while healthy German students displayed enhanced LHPA axis activity and depressiveness, and a lower sense of coherence (Narita et al., 2012; Ullmann et al., 2017). Higher levels of recalled parental overprotection were also associated with anxiety, psychosomatic symptoms, and eating disorders (Ihle et al., 2005; Ullmann et al., 2013). Our current results show that parental overprotection is associated with LHPA axis activation in German natives and Russian-German immigrants. In contrast, we found dampened LHPA axis activity in Jewish subjects with increased parental overprotection but without mental abnormalities. This suggests an inherited allostatic overload of the LHPA axis, which might arise from a gene methylation of the glucocorticoid receptor or an insufficiency of 21-hydroxylase after mental burdening (McEwen, 2005; Speiser et al., 1985; Yehuda et al., 2016). Our data on anti-Semitic experiences as associated with higher levels of maternal overprotection support this hypothesis. We posit that the natural adaptive processes within LHPA axis activity may promote higher individual stress resistance, since it requires a significant stressor to activate steroid release. From a salutogenic perspective, this increased stress resistence may be a survival strategy of children with an anxious overprotective mother. Understanding the molecular mechanisms mediating allostasis may promote new therapeutic pathways and identify anti-vulnerability processes. This knowledge will be tantamount to developing the alternative approaches necessary to clarify the neurobiology and mechanisms relevant to development, diagnosis, and treatment of stress related disorders (Krishnan & Nestler, 2008).

Limitations

The subjects in our study were not screened for clinical diagnosis in respect to ICD-10/DSM 5, and the groups were not assessed specifically for trauma exposure but only in regard to their level of anxiety, depressiveness, and physical complaints. Mother-/parent–child interactions were not observed directly. However, we suggest that parental overprotective child-rearing may be an indirect measure of inhibited individuation. Also, women on oral contraception were not excluded, which may have influenced the steroid profiles. Moreover, measurements of anti-Semitism experiences have not yet been standardized.

Conclusions

We propose considering a successful individuation process and the development of positive parent-child dyads to be key predictors of healthy emotional competence. Frustrated individuation can lead to an allostatic overload. Anxious, mentally-burdened mothers protect their children to preserve their personal containing object, which inhibits individuation and alters the LHPA axis activity. Our data suggest that clinical assessment of hair-steroid concentrations and parenting practices could be useful for diagnosing stress related disorders. Our study will stimulate future research in early childhood development, with special regard to the mother–child relationships and the consequences of overprotective parenting.

Lay summary

Using Jewish, German, and Russian-German subjects, we investigated whether parental rearing behavior influences stress hormone systems. An opposite stress hormone response was found in the Jewish subjects experiencing anti-Semitism, versus their German and Russian-German counterparts. This may indicate a disruption of early mother-child bonding relationships in this population, with potential implications for the development of PTSD.

Acknowledgements

This study was supported by institutional funds from the State University of New York (SUNY) Upstate Medical University. This is a TransCampus project.

Disclosure statement

No potential conflict of interest was reported by the authors.