- Email: [email protected]
Individual differences in emotional expressivity predict oxytocin responses to cortisol administration: Relevance to breast cancer?
Biological Psychology 75 (2007) 119–123 www.elsevier.com/locate/biopsycho
Individual differences in emotional expressivity predict oxytocin responses to cortisol administration: Relevance to breast cancer? Mattie Tops a,b,c,*, Jacobien M. van Peer a,b,d, Jakob Korf b a
Department of Experimental and Work Psychology, University of Groningen, Grote Kruisstraat 2/1, NL-9712 TS Groningen, The Netherlands Department of Psychiatry, University Medical Center Groningen, Hanzeplein 1, P.O. Box 30.001, NL-9700 RB Groningen, The Netherlands c Centre for Child and Family Studies, University of Leiden, P.O. Box 9555, NL-2300 RB Leiden, The Netherlands d Section of Clinical and Health Psychology, University of Leiden, P.O. Box 9555, NL-2300 RB Leiden, The Netherlands
b
Received 28 September 2006; accepted 3 January 2007 Available online 7 January 2007
Abstract Reduced emotional expression has been consistently related to susceptibility or fast progression of breast cancer. Breast cancer development and reduced emotional expression have both been related to rejection- and separation-related conditions. The neuropeptide oxytocin is low in response to rejection or separation. Recent results suggest that oxytocin may protect against the development of breast cancer and slow its progression. In the present study, we investigated if individual differences in emotional expressivity relate to basal or cortisol-stimulated plasma oxytocin. Healthy female subjects were treated with placebo or 35 mg of cortisol orally in a double-blind within-subject study. Seventy minutes later, blood was sampled for determination of oxytocin and cortisol levels. We found an interaction between treatment condition and Emotional Expression-out scores: after cortisol treatment, oxytocin levels increased proportional to Emotion Expression-out score. These preliminary findings provide a potential mechanism for associations in the literature between emotional expressive behavior and breast cancer. # 2007 Elsevier B.V. All rights reserved. Keywords: Emotional suppression; Emotional expression; Cortisol; Oxytocin; Breast cancer; Fear of rejection
1. Introduction Reduced emotional expression has been consistently related to susceptibility or fast progression of breast cancer, although no physiological mechanism is known that could explain this association (McKenna et al., 1999). Breast cancer development and reduced emotional expression have both been related to rejection- and separation-related conditions (McKenna et al., 1999; Suslow et al., 2000). Oxytocin, a neuropeptide involved in attachment mechanisms, is low in response to rejection or separation (Carter, 1998; Panksepp et al., 1997). Recent results suggest that oxytocin may protect against the development of breast cancer and slow its progression (Carrera et al., 2004; Cassoni et al., 2004). Oxytocin interacts with glucocorticoids such as cortisol (Carter, 1998; Liberzon and Young, 1997), and fear of rejection is a reliable inducer of cortisol responses in * Corresponding author at: Centre for Child and Family Studies, University of Leiden, P.O. Box 9555, NL-2300 RB Leiden, The Netherlands. Tel.: +31 71 5273434; fax: +31 71 5273945. E-mail address: [email protected] (M. Tops). 0301-0511/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.biopsycho.2007.01.001
humans (Dickerson and Kemeny, 2004). In the present study, we hypothesized that, in healthy women, individual differences in emotional expressivity relate to basal or cortisol-stimulated plasma oxytocin. A relationship between emotional expressivity and oxytocin levels would provide a potential mechanism for associations between emotional expressive behavior and breast cancer. In the remainder of this section, we will elaborate on this line of thought (see Fig. 1), discus evidence in support of it, and present hypotheses that follow from it. There are indications that attachment mechanisms, such as fear of rejection, may moderate the association between reduced emotional expression and breast cancer. A metaanalysis of the psychosocial factors related to the development of breast cancer found the largest significant moderate effect size for emotional suppression/alexithymia, followed by separation/loss, stressful live events and conflict-avoidant personality style (McKenna et al., 1999). Conflict avoidance entails the maintenance of harmony and avoidance of rejection via inhibition of expression of feelings and emotions (Barry and Mizrahi, 2005), and indeed, fear of rejection seems to be an important suppressive regulator of emotional expression
120
M. Tops et al. / Biological Psychology 75 (2007) 119–123
Fig. 1. Depiction of the relationships between variables discussed in Section 1. In addition to the depicted relationships, the psychosocial factors on the right of the figure may influence, and be influenced by, the interactions between cortisol and oxytocin. This study investigated if individual differences in emotional expressivity in healthy women relate to basal or cortisol-induced plasma oxytocin levels, because such relationships may explain reported associations between low emotional expressivity and development of breast cancer.
(Suslow et al., 2000). Interestingly, insecure attachment styles have been related to reduced emotional expression, especially in adults not involved in a relationship (Kerr et al., 2003), and breast cancer patients scored higher on insecure attachment and emotional suppression compared to controls (Tacon et al., 2001). So it seems that psychosocial factors most likely to relate to breast cancer involve attachment issues such as separation, rejection and loss, and an inhibitory avoidant way of coping with such stressors. Research on the neurobiology of attachment in mammals suggests that the neuropeptide oxytocin is a modulator of attachment processes. Oxytocin is low in response to rejection or separation, and oxytocin administration reduces separation anxiety (Carter, 1998; Panksepp et al., 1997). The oxytocin system operates in parallel with known stress response systems both to bring about ‘‘calm and connection’’ effects in response to positive social cues, and to inhibit sympathetic and HPA activity and promote parasympathetic cardiac control after stress (Uvna¨s-Moberg et al., 2005). Especially in anxietyprovoking situations, oxytocin reduces the behavioral inhibition in social interactions that leads to emotional suppression (Carter, 1998; McCarthy, 1995; Panksepp et al., 1997). Oxytocin may interact with glucocorticoids to influence behavior (Carter, 1998; Liberzon and Young, 1997). Steroid hormones including glucocorticoids can influence the synthesis of, release of, and/or receptors for neuropeptides. Kalin et al. (1985) found that administration of the synthetic glucocorticoid dexamethasone to rhesus monkeys resulted in increases in plasma oxytocin levels. Modulation of oxytocin activity by glucocorticoids may explain facilitatory effects of stress and glucocorticoids on attachment. Studies on human and animal behavior consistently implicate stress and glucocorticoids such as cortisol and corticosterone in the formation of attachments. Stress or corticosterone injection facilitates pair bonding in prairie voles (DeVries et al., 1996). Interestingly, a metaanalysis showed that fear of negative social evaluation (i.e. fear of rejection) is one of the most reliable psychosocial elicitors of cortisol responses in laboratory studies of humans (Dickerson and Kemeny, 2004). As oxytocin is believed to modulate
separation and rejection-related anxiety, this suggests that individual differences in oxytocin modulation by cortisol may translate behaviorally into individual differences in fear of rejection and emotional expression. It has recently been found that oxytocin may protect against the development of breast cancer and slow progression (Carrera et al., 2004; Cassoni et al., 2004; Murrell, 1995). As discussed above, separation/loss experience relates to low levels of oxytocin and may be associated with the development of breast cancer (McKenna et al., 1999). This points to a possible mechanism through which emotional suppression, if related to low oxytocin levels, could be associated with the risk of breast cancer (see Fig. 1). In the present study, emotional expression was measured with the Emotional Expression and Control (EEC) questionnaire. This questionnaire was constructed by Bleiker et al. (1993) from items of previous questionnaires used in breast cancer research, and subsequently validated and used in research (Bleiker et al., 1993). As we propose that oxytocin may mediate the protective effects of emotional expression, we expected higher oxytocin levels, and/or larger cortisol-induced oxytocin increases, in healthy female subjects who selfreported higher levels of emotional expressivity. 2. Methods 2.1. Participants Eighteen healthy female paid volunteers aged 30–52 (mean age = 41) were enlisted by advertisement in a local newspaper. Each participant passed a health screening based on self-report. Inclusion criteria included regularly cycling or using oral contraception, and no personal history of psychiatric, metabolic, or neurological disorders or substance abuse. Volunteers who reported noxious health behaviors (drug abuse including excessive alcohol, smoking and caffeine, and abnormal sleeping habits, e.g. too little sleep), chronic health problems or psychopathology were excluded from the study. Our subjects self-reported the phase of their menstrual cycle (follicular, midcycle, or luteal) based on days since onset of their last menstrual period, and active contraceptive use (six subjects). All participants read and signed an informed consent statement approved by the Medical Ethical Committee of the University Medical Center of Groningen.
2.2. Procedure We used a within-subjects double-blind design. Placebo and treatment sessions, the order of which was counterbalanced, were separated by approximately 1 week. The participants arrived between 9:00 and 10:30. In the experimental condition participants received a capsule containing cortisol (35 mg of hydrocortisone), whereas in the control condition they received a placebo (avicel capsule) double-blind orally. Participants filled out the questionnaires and were then allowed to read while they waited 70 min for bloodsampling. We found central effects using this dose of cortisol and timing in several other studies (e.g. Tops et al., 2006). Next, participants performed memory tasks for 1 h, the results of which will be presented elsewhere.
2.3. Emotional Expression and Control (EEC) Emotional expression was measured with the Emotional Expression and Control (EEC) questionnaire. This questionnaire was constructed by Bleiker et al. (1993) of items from previous questionnaires used in cancer research, and subsequently validated and used in research. It contains the sub-scales Emotional Control (six items), Emotional Expression-out (Emotion-out, six items,
M. Tops et al. / Biological Psychology 75 (2007) 119–123
121
e.g. ‘‘when I feel unhappy or miserable, I say what I feel’’), and Emotional Expression-in (Emotion-in, six items, e.g. ‘‘when I feel unhappy or miserable, I hide my unhappiness’’). Answers were given on four-point scales ranging from ‘‘almost never’’ (scored 1) to ‘‘almost always’’ (scored 4), and scores were summed to obtain scale scores between 6 and 24.
2.4. Plasma hormone levels The blood samples were collected in ice-chilled tubes. After centrifugation, the plasma was removed and the samples were stored at 80 8C until analysis. Plasma concentrations of cortisol were determined using an in-house radioimmunoassay (Pratt, 1978). These analyses were performed at the University Medical Center of Groningen. Before oxytocin was analyzed, plasma was extracted with acetone (Guarantee Reagent (GR); Merck, Darmstadt, Germany) and petroleum benzene (GR; boiling point range 40–60 8C; Merck) with a recovery of 92%. Oxytocin levels were measured by radioimmunoassay at the Swedish University of Agricultural Sciences in Uppsala, Sweden, at the Department of Animal Physiology. The antibody KA19 was used for the analysis (Milab, Sweden). The limit of detection was 4.68 pmol/l and interassay coefficients of variation low 22.06% CV = 9.57; medium 37.83% CV = 8.81; high 529.8% CV = 6.81; intrassay % CV < 10, conc. = 20.90–1026 (Stock and Uvna¨s-Moberg, 1988).
3. Results 3.1. Emotion-in and Emotion-out Mean Emotion-in score was 14.80, S.D. = 3.34; mean Emotion-out was 13.50, S.D. = 2.45. Emotion-in and Emotionout tended to correlate negatively (r = .42, p = .085). 3.2. Hormones Levels and response of cortisol and oxytocin were not influenced by age, menstrual cycle phase or oral contraceptive use. Cortisol administration increased plasma cortisol levels (placebo: M = 204 nmol/l, S.D. = 58; treatment: M = 1283, S.D. = 413; F(1, 17) = 138.85, p = .00001). Oxytocin levels in the placebo condition correlated positively with levels in the treatment condition (r = .69, p = .001). General Linear Model (GLM) analyses for repeated measures with Treatment condition as the within-subjects factor, and Emotionin and Emotion-out scores as continuous covariates, showed that oxytocin levels increased from 49.7 pmol/l (S.D. = 9.6) with placebo to 54.6 pmol/l (S.D. = 11) with cortisol administration (F(1, 16) = 4.27, p = .055). Importantly, as is shown in Figs. 2 and 3, oxytocin only increased in subjects scoring higher on Emotion-out (Treatment Emotion-out: F(1, 16) = 6.78, p = .019). Emotion-out scores were significantly correlated to oxytocin levels in the cortisol treatment condition (r = .58, p = .011) and to the treatment-induced increase in oxytocin levels (r = .55, p = .019).
Fig. 2. Mean plasma levels of oxytocin and standard errors of the mean as a function of treatment condition and scores on Emotion-out. For illustrative purposes only, subjects are divided into an Emotion-out low-scoring and a highscoring group based on median split. Subjects scoring high on Emotion-out display an increase in oxytocin levels after cortisol treatment.
important finding, as it may provide a mechanism behind reports of a relationship between emotional suppression and breast cancer. The breasts contain oxytocin target tissues that physiologically express oxytocin receptors, and it has recently become clear that in tumors, oxytocin regulates growth through the oxytocin receptor (Cassoni et al., 2004). Recent results in rats confirm involvement of oxytocin in suppressing breast carcinogenesis (Carrera et al., 2004). In Section 1, we discussed that separation/loss is related to low levels of oxytocin and prospectively to the development of breast cancer. This points to a possible protective mechanism against breast cancer involving relations between affiliative behavior involving emotional expression, and higher oxytocin levels. Indeed, a recent meta-analysis indicated modest associations between emotional suppression/alexithymia, separation/loss, conflictavoidant personality style and the development of breast cancer (McKenna et al., 1999). Similarly, a review concluded that emotion suppression or alexithymia were the strongest predictors of breast cancer especially in younger women (Butow et al., 2000). We did not find relations of oxytocin with Emotion-in, a scale with similar, but oppositely worded items compared to, and negatively correlated with, Emotion-out (see Section 2). Although both scales have been related to breast cancer, we speculate that they may have different sensitivities by covering different ranges of a single dimension of emotional
4. Discussion We found that cortisol treatment and Emotion-out scores interacted in determining plasma oxytocin levels. After treatment with cortisol, oxytocin levels increased proportional to the Emotion-out score. The higher oxytocinergic response in subjects who report a high level of emotional expression is an
Fig. 3. Scatterplot displaying the relationship between Emotion-out scores and cortisol-induced increase in plasma oxytocin levels (r = .55, p = .019).
122
M. Tops et al. / Biological Psychology 75 (2007) 119–123
expressivity; beneficial effects of expression of emotions may be more easily detected in mentally healthy populations, whereas harmful effects of inhibition of emotions may be more easily detected in non-healthy populations (Takagi and Ohira, 2004). Several aspects define this study as preliminary. The number of subjects was small which means that the power to detect main effects of Emotion-In and Emotion-out was low. Also a direct psychological effect of plasma oxytocin cannot be assumed because peripheral and central release of oxytocin are not always coordinated. However, previous animal and some human studies have shown plasma oxytocin levels to be positively associated with attachment behaviors (Light et al., 2005; Uvna¨s-Moberg et al., 1999, 2005). Another limitation was that we relied on a single sample for hormone determination per session, although the high correlation between the oxytocin levels of the two conditions suggests good reliability of the oxytocin measurement. Finally, we included subjects in various phases of their menstrual cycle. It is unclear whether this affected results. Consistent with larger studies employing between-subjects comparisons, we did not observe any relationships between menstrual cycle phase or oral contraceptive use and oxytocin levels or oxytocin responses (Light et al., 2005). In contrast, studies employing a within-subject design report lower oxytocin levels in the luteal phase (Salonia et al., 2005), which suggests that menstrual cycle phase variations in oxytocin levels are small and might be easily masked by individual differences (Light et al., 2005). As oxytocin has anxiolytic effects (Uvna¨s-Moberg et al., 2005), the cortisol-induced increases in oxytocin are worth further study in the context of a recent report that cortisol administration protects against the fear arousing effect of phobia-related stimuli in subjects with social phobia or spider phobia (Soravia et al., 2006), and another study that found that patients that received cortisol during the perioperative period of cardiac surgery had lower intensity of chronic stress and posttraumatic stress disorder symptoms at 6 months after surgery (Schelling et al., 2004). As concluded by McKenna et al. (1999) in their metaanalysis of the psychosocial factors influencing the development of breast cancer, future research should specify pathways through which biological phenomena affect the development of cancer and, after establishing the existence of such mechanisms, work back to demonstrate how psychosocial factors may influence them. Moderating and/or mediating relationships may be the key to a more complete understanding. Future research should adopt theoretical models and methodologies that consider and examine the interplay of hormonal, physiological, and psychological responses in the development of breast disease. Furthermore, this research would be best focused on those factors for which evidence was found in the metaanalysis, namely, emotional suppression, separation/loss, stressful life events, and conflict-avoidance personality style (McKenna et al., 1999), factors that we argued may involve attachment mechanisms. Although the present results are preliminary, we think that the approach and results of the
present study fulfill many of the above recommendations, and hence may warrant further research. Acknowledgements Parts of this research is supported by the Netherlands Concerted Research Action ‘Fatigue at work’ of the Netherlands Research Organization (NWO) (580-02.100D) and by a grant (# 9910) of the Gratama Foundation. References Barry, D.T., Mizrahi, T.C., 2005. Guarded self-disclosure predicts psychological distress and willingness to use psychological services among Asian immigrants in the United States. Journal of Nervous and Mental Disorders 193 (8), 535–539. Bleiker, E.M.A., van der Ploeg, H.M., Hendriks, J.H.C.L., Leer, J.-W.H., Kleijn, W.C., 1993. Rationality, emotional expression and control: psychometric characteristics of a questionnaire for research in psycho-oncology. Journal of Psychosomatic Research 37, 861–872. Butow, P.N., Hiller, J.E., Price, M.A., Thackway, S.V., Kricker, A., Tennant, C.C., 2000. Epidemiological evidence for a relationship between life events, coping style, and personality factors in the development of breast cancer. Journal of Psychosomatic Research 49, 169–181. Carrera, M.P., Ramirex-Exposito, M.J., Valenzuela, M.T., Garcia, M.J., Mayas, M.D., Martinez-Martos, J.M., 2004. Serum oxytocinase activity is related to tumor growth parameters in N-methyl nitrosourea induced rat breast cancer. Life Sciences 75, 1369–1377. Carter, C.S., 1998. Neuroendocrine perspectives on social attachment and love. Psychoneuroendocrinology 23, 779–818. Cassoni, P., Sapino, A., Marrocco, T., Chini, B., Bussolati, G., 2004. Oxytocin and oxytocin receptors in cancer cells and proliferation. Journal of Neuroendocrinology 16, 362–364. DeVries, A.C., DeVries, M.B., Taymans, S.E., Carter, C.S., 1996. The effects of stress on social preferences are sexually dimorphic in prairie voles. Proceedings of the National Academy of Sciences of the United States of America 93, 11980–11984. Dickerson, S.S., Kemeny, M.E., 2004. Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. Psychological Bulletin 130, 355–391. Kalin, N.H., Gibbs, D.M., Barksdale, C.M., Shelton, S.E., Carnes, M., 1985. Behavioral stress decreases plasma oxytocin concentrations in primates. Life Sciences 36, 1275–1280. Kerr, S.L., Melley, A.M., Travea, L., Pole, M., 2003. The relationship of emotional expression and experience to adult attachment style. Individual Differences Research 1 (2), 108–123. Liberzon, I., Young, E.A., 1997. Effects of stress and glucocorticoids on CNS oxytocin receptor binding. Psychoneuroendocrinology 22, 411–422. Light, K.L., Grewen, K.M., Amico, J.A., 2005. More frequent partner hugs and higher oxytocin levels are linked to lower blood pressure and heart rate in premenopausal women. Biological Psychology 69, 5–21. McCarthy, M.M., 1995. Estrogen modulation of oxytocin and its relation to behavior. Advances in Experimental Medical Biology 395, 235–245. McKenna, M., Zevon, M., Corn, B., Rounds, J., 1999. Psychosocial factors and the development of breast cancer: a meta-analysis. Health Psychology 18, 520–531. Murrell, T.G., 1995. The potential for oxytocin to prevent breast cancer: a hypothesis. Breast Cancer Research and Treatment 35, 225–229. Panksepp, J., Nelson, E., Bekkedal, M., 1997. Brain systems for the mediation of social separation-distress and social-reward. Annals of the New York Academy of Sciences 807, 78–100. Pratt, J.J., 1978. Steroid immunoassay in clinical chemistry. Clinical Chemistry 24, 1869–1890. Salonia, A., Nappi, R.E., Pontillo, M., Daverio, R., Smeraldi, A., Briganti, A., Fabbri, F., Zanni, G., Rigatti, P., Montorsi, F., 2005. Menstrual cycle-related
M. Tops et al. / Biological Psychology 75 (2007) 119–123 changes in plasma oxytocin are relevant to normal sexual function in healthy women. Hormones and Behaviour 47, 164–169. Schelling, G., Kilger, E., Roozendaal, B., de Quervain, D.J.-F., Briegel, J., Dagge, A., Rothenhausler, H.B., Krauseneck, T., Nollert, G., Kapfhammer, H.P., 2004. Stress doses of hydrocortisone, traumatic memories, and symptoms of posttraumatic stress disorder in patients after cardiac surgery: a randomized study. Biological Psychology 55, 627–633. Soravia, L.M., Heinrichs, M., Aerni, A., Maroni, C., Schelling, G., Ehlert, U., Roozendaal, B., de Quervain, D.J.-F., 2006. Glucocorticoids reduce phobic fear in humans. Proceedings of the National Academy of Sciences of the United States of America 103, 5585–5590. Stock, S., Uvna¨s-Moberg, K., 1988. Increased plasma levels of oxytocin in response to afferent electrical stimulation in sciatic and vagal nerves and in response to touch and pinch in anaesthetized rats. Acta Physiologica Scandinavia 132, 29–34. Suslow, T., Donges, U.-S., Kersting, A., Arolt, V., 2000. 20-Item Toronto Alexithymia Scale: do difficulties describing feelings assess proneness to
123
shame instead of difficulties symbolizing emotions? Scandinavian Journal of Psychology 41, 329–334. Tacon, A.M., Caldera, Y.M., Bell, N.J., 2001. Attachment style, emotional control, and breast cancer. Family Systems and Health 19, 319–326. Takagi, S., Ohira, H., 2004. Effects of expression and inhibition of negative emotions on health, mood states, and salivary secretory immunoglobulin A in Japanese mildly depressed undergraduates. Perceptual and Motor Skills 98, 1187–1198. Tops, M., van Peer, J.M., Wester, E.A., Wijers, A.A., Korf, J., 2006. Statedependent regulation of cortical activity by cortisol: an EEG study. Neuroscience Letters 404, 39–43. Uvna¨s-Moberg, K., Bjo¨rkstrand, E., Hillegaart, V., Ahlenius, S., 1999. Oxytocin as a possible mediator of SSRI-induced antidepressant effects. Psychopharmacology 142, 95–101. Uvna¨s-Moberg, K., Arn, I., Magnusson, D., 2005. The psychobiology of emotion: the role of the oxytocinergic system. International Journal of Behavioral Medicine 12, 59–65.
Individual differences in emotional expressivity predict oxytocin responses to cortisol administration: Relevance to breast cancer? Mattie Tops a,b,c,*, Jacobien M. van Peer a,b,d, Jakob Korf b a
Department of Experimental and Work Psychology, University of Groningen, Grote Kruisstraat 2/1, NL-9712 TS Groningen, The Netherlands Department of Psychiatry, University Medical Center Groningen, Hanzeplein 1, P.O. Box 30.001, NL-9700 RB Groningen, The Netherlands c Centre for Child and Family Studies, University of Leiden, P.O. Box 9555, NL-2300 RB Leiden, The Netherlands d Section of Clinical and Health Psychology, University of Leiden, P.O. Box 9555, NL-2300 RB Leiden, The Netherlands
b
Received 28 September 2006; accepted 3 January 2007 Available online 7 January 2007
Abstract Reduced emotional expression has been consistently related to susceptibility or fast progression of breast cancer. Breast cancer development and reduced emotional expression have both been related to rejection- and separation-related conditions. The neuropeptide oxytocin is low in response to rejection or separation. Recent results suggest that oxytocin may protect against the development of breast cancer and slow its progression. In the present study, we investigated if individual differences in emotional expressivity relate to basal or cortisol-stimulated plasma oxytocin. Healthy female subjects were treated with placebo or 35 mg of cortisol orally in a double-blind within-subject study. Seventy minutes later, blood was sampled for determination of oxytocin and cortisol levels. We found an interaction between treatment condition and Emotional Expression-out scores: after cortisol treatment, oxytocin levels increased proportional to Emotion Expression-out score. These preliminary findings provide a potential mechanism for associations in the literature between emotional expressive behavior and breast cancer. # 2007 Elsevier B.V. All rights reserved. Keywords: Emotional suppression; Emotional expression; Cortisol; Oxytocin; Breast cancer; Fear of rejection
1. Introduction Reduced emotional expression has been consistently related to susceptibility or fast progression of breast cancer, although no physiological mechanism is known that could explain this association (McKenna et al., 1999). Breast cancer development and reduced emotional expression have both been related to rejection- and separation-related conditions (McKenna et al., 1999; Suslow et al., 2000). Oxytocin, a neuropeptide involved in attachment mechanisms, is low in response to rejection or separation (Carter, 1998; Panksepp et al., 1997). Recent results suggest that oxytocin may protect against the development of breast cancer and slow its progression (Carrera et al., 2004; Cassoni et al., 2004). Oxytocin interacts with glucocorticoids such as cortisol (Carter, 1998; Liberzon and Young, 1997), and fear of rejection is a reliable inducer of cortisol responses in * Corresponding author at: Centre for Child and Family Studies, University of Leiden, P.O. Box 9555, NL-2300 RB Leiden, The Netherlands. Tel.: +31 71 5273434; fax: +31 71 5273945. E-mail address: [email protected] (M. Tops). 0301-0511/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.biopsycho.2007.01.001
humans (Dickerson and Kemeny, 2004). In the present study, we hypothesized that, in healthy women, individual differences in emotional expressivity relate to basal or cortisol-stimulated plasma oxytocin. A relationship between emotional expressivity and oxytocin levels would provide a potential mechanism for associations between emotional expressive behavior and breast cancer. In the remainder of this section, we will elaborate on this line of thought (see Fig. 1), discus evidence in support of it, and present hypotheses that follow from it. There are indications that attachment mechanisms, such as fear of rejection, may moderate the association between reduced emotional expression and breast cancer. A metaanalysis of the psychosocial factors related to the development of breast cancer found the largest significant moderate effect size for emotional suppression/alexithymia, followed by separation/loss, stressful live events and conflict-avoidant personality style (McKenna et al., 1999). Conflict avoidance entails the maintenance of harmony and avoidance of rejection via inhibition of expression of feelings and emotions (Barry and Mizrahi, 2005), and indeed, fear of rejection seems to be an important suppressive regulator of emotional expression
120
M. Tops et al. / Biological Psychology 75 (2007) 119–123
Fig. 1. Depiction of the relationships between variables discussed in Section 1. In addition to the depicted relationships, the psychosocial factors on the right of the figure may influence, and be influenced by, the interactions between cortisol and oxytocin. This study investigated if individual differences in emotional expressivity in healthy women relate to basal or cortisol-induced plasma oxytocin levels, because such relationships may explain reported associations between low emotional expressivity and development of breast cancer.
(Suslow et al., 2000). Interestingly, insecure attachment styles have been related to reduced emotional expression, especially in adults not involved in a relationship (Kerr et al., 2003), and breast cancer patients scored higher on insecure attachment and emotional suppression compared to controls (Tacon et al., 2001). So it seems that psychosocial factors most likely to relate to breast cancer involve attachment issues such as separation, rejection and loss, and an inhibitory avoidant way of coping with such stressors. Research on the neurobiology of attachment in mammals suggests that the neuropeptide oxytocin is a modulator of attachment processes. Oxytocin is low in response to rejection or separation, and oxytocin administration reduces separation anxiety (Carter, 1998; Panksepp et al., 1997). The oxytocin system operates in parallel with known stress response systems both to bring about ‘‘calm and connection’’ effects in response to positive social cues, and to inhibit sympathetic and HPA activity and promote parasympathetic cardiac control after stress (Uvna¨s-Moberg et al., 2005). Especially in anxietyprovoking situations, oxytocin reduces the behavioral inhibition in social interactions that leads to emotional suppression (Carter, 1998; McCarthy, 1995; Panksepp et al., 1997). Oxytocin may interact with glucocorticoids to influence behavior (Carter, 1998; Liberzon and Young, 1997). Steroid hormones including glucocorticoids can influence the synthesis of, release of, and/or receptors for neuropeptides. Kalin et al. (1985) found that administration of the synthetic glucocorticoid dexamethasone to rhesus monkeys resulted in increases in plasma oxytocin levels. Modulation of oxytocin activity by glucocorticoids may explain facilitatory effects of stress and glucocorticoids on attachment. Studies on human and animal behavior consistently implicate stress and glucocorticoids such as cortisol and corticosterone in the formation of attachments. Stress or corticosterone injection facilitates pair bonding in prairie voles (DeVries et al., 1996). Interestingly, a metaanalysis showed that fear of negative social evaluation (i.e. fear of rejection) is one of the most reliable psychosocial elicitors of cortisol responses in laboratory studies of humans (Dickerson and Kemeny, 2004). As oxytocin is believed to modulate
separation and rejection-related anxiety, this suggests that individual differences in oxytocin modulation by cortisol may translate behaviorally into individual differences in fear of rejection and emotional expression. It has recently been found that oxytocin may protect against the development of breast cancer and slow progression (Carrera et al., 2004; Cassoni et al., 2004; Murrell, 1995). As discussed above, separation/loss experience relates to low levels of oxytocin and may be associated with the development of breast cancer (McKenna et al., 1999). This points to a possible mechanism through which emotional suppression, if related to low oxytocin levels, could be associated with the risk of breast cancer (see Fig. 1). In the present study, emotional expression was measured with the Emotional Expression and Control (EEC) questionnaire. This questionnaire was constructed by Bleiker et al. (1993) from items of previous questionnaires used in breast cancer research, and subsequently validated and used in research (Bleiker et al., 1993). As we propose that oxytocin may mediate the protective effects of emotional expression, we expected higher oxytocin levels, and/or larger cortisol-induced oxytocin increases, in healthy female subjects who selfreported higher levels of emotional expressivity. 2. Methods 2.1. Participants Eighteen healthy female paid volunteers aged 30–52 (mean age = 41) were enlisted by advertisement in a local newspaper. Each participant passed a health screening based on self-report. Inclusion criteria included regularly cycling or using oral contraception, and no personal history of psychiatric, metabolic, or neurological disorders or substance abuse. Volunteers who reported noxious health behaviors (drug abuse including excessive alcohol, smoking and caffeine, and abnormal sleeping habits, e.g. too little sleep), chronic health problems or psychopathology were excluded from the study. Our subjects self-reported the phase of their menstrual cycle (follicular, midcycle, or luteal) based on days since onset of their last menstrual period, and active contraceptive use (six subjects). All participants read and signed an informed consent statement approved by the Medical Ethical Committee of the University Medical Center of Groningen.
2.2. Procedure We used a within-subjects double-blind design. Placebo and treatment sessions, the order of which was counterbalanced, were separated by approximately 1 week. The participants arrived between 9:00 and 10:30. In the experimental condition participants received a capsule containing cortisol (35 mg of hydrocortisone), whereas in the control condition they received a placebo (avicel capsule) double-blind orally. Participants filled out the questionnaires and were then allowed to read while they waited 70 min for bloodsampling. We found central effects using this dose of cortisol and timing in several other studies (e.g. Tops et al., 2006). Next, participants performed memory tasks for 1 h, the results of which will be presented elsewhere.
2.3. Emotional Expression and Control (EEC) Emotional expression was measured with the Emotional Expression and Control (EEC) questionnaire. This questionnaire was constructed by Bleiker et al. (1993) of items from previous questionnaires used in cancer research, and subsequently validated and used in research. It contains the sub-scales Emotional Control (six items), Emotional Expression-out (Emotion-out, six items,
M. Tops et al. / Biological Psychology 75 (2007) 119–123
121
e.g. ‘‘when I feel unhappy or miserable, I say what I feel’’), and Emotional Expression-in (Emotion-in, six items, e.g. ‘‘when I feel unhappy or miserable, I hide my unhappiness’’). Answers were given on four-point scales ranging from ‘‘almost never’’ (scored 1) to ‘‘almost always’’ (scored 4), and scores were summed to obtain scale scores between 6 and 24.
2.4. Plasma hormone levels The blood samples were collected in ice-chilled tubes. After centrifugation, the plasma was removed and the samples were stored at 80 8C until analysis. Plasma concentrations of cortisol were determined using an in-house radioimmunoassay (Pratt, 1978). These analyses were performed at the University Medical Center of Groningen. Before oxytocin was analyzed, plasma was extracted with acetone (Guarantee Reagent (GR); Merck, Darmstadt, Germany) and petroleum benzene (GR; boiling point range 40–60 8C; Merck) with a recovery of 92%. Oxytocin levels were measured by radioimmunoassay at the Swedish University of Agricultural Sciences in Uppsala, Sweden, at the Department of Animal Physiology. The antibody KA19 was used for the analysis (Milab, Sweden). The limit of detection was 4.68 pmol/l and interassay coefficients of variation low 22.06% CV = 9.57; medium 37.83% CV = 8.81; high 529.8% CV = 6.81; intrassay % CV < 10, conc. = 20.90–1026 (Stock and Uvna¨s-Moberg, 1988).
3. Results 3.1. Emotion-in and Emotion-out Mean Emotion-in score was 14.80, S.D. = 3.34; mean Emotion-out was 13.50, S.D. = 2.45. Emotion-in and Emotionout tended to correlate negatively (r = .42, p = .085). 3.2. Hormones Levels and response of cortisol and oxytocin were not influenced by age, menstrual cycle phase or oral contraceptive use. Cortisol administration increased plasma cortisol levels (placebo: M = 204 nmol/l, S.D. = 58; treatment: M = 1283, S.D. = 413; F(1, 17) = 138.85, p = .00001). Oxytocin levels in the placebo condition correlated positively with levels in the treatment condition (r = .69, p = .001). General Linear Model (GLM) analyses for repeated measures with Treatment condition as the within-subjects factor, and Emotionin and Emotion-out scores as continuous covariates, showed that oxytocin levels increased from 49.7 pmol/l (S.D. = 9.6) with placebo to 54.6 pmol/l (S.D. = 11) with cortisol administration (F(1, 16) = 4.27, p = .055). Importantly, as is shown in Figs. 2 and 3, oxytocin only increased in subjects scoring higher on Emotion-out (Treatment Emotion-out: F(1, 16) = 6.78, p = .019). Emotion-out scores were significantly correlated to oxytocin levels in the cortisol treatment condition (r = .58, p = .011) and to the treatment-induced increase in oxytocin levels (r = .55, p = .019).
Fig. 2. Mean plasma levels of oxytocin and standard errors of the mean as a function of treatment condition and scores on Emotion-out. For illustrative purposes only, subjects are divided into an Emotion-out low-scoring and a highscoring group based on median split. Subjects scoring high on Emotion-out display an increase in oxytocin levels after cortisol treatment.
important finding, as it may provide a mechanism behind reports of a relationship between emotional suppression and breast cancer. The breasts contain oxytocin target tissues that physiologically express oxytocin receptors, and it has recently become clear that in tumors, oxytocin regulates growth through the oxytocin receptor (Cassoni et al., 2004). Recent results in rats confirm involvement of oxytocin in suppressing breast carcinogenesis (Carrera et al., 2004). In Section 1, we discussed that separation/loss is related to low levels of oxytocin and prospectively to the development of breast cancer. This points to a possible protective mechanism against breast cancer involving relations between affiliative behavior involving emotional expression, and higher oxytocin levels. Indeed, a recent meta-analysis indicated modest associations between emotional suppression/alexithymia, separation/loss, conflictavoidant personality style and the development of breast cancer (McKenna et al., 1999). Similarly, a review concluded that emotion suppression or alexithymia were the strongest predictors of breast cancer especially in younger women (Butow et al., 2000). We did not find relations of oxytocin with Emotion-in, a scale with similar, but oppositely worded items compared to, and negatively correlated with, Emotion-out (see Section 2). Although both scales have been related to breast cancer, we speculate that they may have different sensitivities by covering different ranges of a single dimension of emotional
4. Discussion We found that cortisol treatment and Emotion-out scores interacted in determining plasma oxytocin levels. After treatment with cortisol, oxytocin levels increased proportional to the Emotion-out score. The higher oxytocinergic response in subjects who report a high level of emotional expression is an
Fig. 3. Scatterplot displaying the relationship between Emotion-out scores and cortisol-induced increase in plasma oxytocin levels (r = .55, p = .019).
122
M. Tops et al. / Biological Psychology 75 (2007) 119–123
expressivity; beneficial effects of expression of emotions may be more easily detected in mentally healthy populations, whereas harmful effects of inhibition of emotions may be more easily detected in non-healthy populations (Takagi and Ohira, 2004). Several aspects define this study as preliminary. The number of subjects was small which means that the power to detect main effects of Emotion-In and Emotion-out was low. Also a direct psychological effect of plasma oxytocin cannot be assumed because peripheral and central release of oxytocin are not always coordinated. However, previous animal and some human studies have shown plasma oxytocin levels to be positively associated with attachment behaviors (Light et al., 2005; Uvna¨s-Moberg et al., 1999, 2005). Another limitation was that we relied on a single sample for hormone determination per session, although the high correlation between the oxytocin levels of the two conditions suggests good reliability of the oxytocin measurement. Finally, we included subjects in various phases of their menstrual cycle. It is unclear whether this affected results. Consistent with larger studies employing between-subjects comparisons, we did not observe any relationships between menstrual cycle phase or oral contraceptive use and oxytocin levels or oxytocin responses (Light et al., 2005). In contrast, studies employing a within-subject design report lower oxytocin levels in the luteal phase (Salonia et al., 2005), which suggests that menstrual cycle phase variations in oxytocin levels are small and might be easily masked by individual differences (Light et al., 2005). As oxytocin has anxiolytic effects (Uvna¨s-Moberg et al., 2005), the cortisol-induced increases in oxytocin are worth further study in the context of a recent report that cortisol administration protects against the fear arousing effect of phobia-related stimuli in subjects with social phobia or spider phobia (Soravia et al., 2006), and another study that found that patients that received cortisol during the perioperative period of cardiac surgery had lower intensity of chronic stress and posttraumatic stress disorder symptoms at 6 months after surgery (Schelling et al., 2004). As concluded by McKenna et al. (1999) in their metaanalysis of the psychosocial factors influencing the development of breast cancer, future research should specify pathways through which biological phenomena affect the development of cancer and, after establishing the existence of such mechanisms, work back to demonstrate how psychosocial factors may influence them. Moderating and/or mediating relationships may be the key to a more complete understanding. Future research should adopt theoretical models and methodologies that consider and examine the interplay of hormonal, physiological, and psychological responses in the development of breast disease. Furthermore, this research would be best focused on those factors for which evidence was found in the metaanalysis, namely, emotional suppression, separation/loss, stressful life events, and conflict-avoidance personality style (McKenna et al., 1999), factors that we argued may involve attachment mechanisms. Although the present results are preliminary, we think that the approach and results of the
present study fulfill many of the above recommendations, and hence may warrant further research. Acknowledgements Parts of this research is supported by the Netherlands Concerted Research Action ‘Fatigue at work’ of the Netherlands Research Organization (NWO) (580-02.100D) and by a grant (# 9910) of the Gratama Foundation. References Barry, D.T., Mizrahi, T.C., 2005. Guarded self-disclosure predicts psychological distress and willingness to use psychological services among Asian immigrants in the United States. Journal of Nervous and Mental Disorders 193 (8), 535–539. Bleiker, E.M.A., van der Ploeg, H.M., Hendriks, J.H.C.L., Leer, J.-W.H., Kleijn, W.C., 1993. Rationality, emotional expression and control: psychometric characteristics of a questionnaire for research in psycho-oncology. Journal of Psychosomatic Research 37, 861–872. Butow, P.N., Hiller, J.E., Price, M.A., Thackway, S.V., Kricker, A., Tennant, C.C., 2000. Epidemiological evidence for a relationship between life events, coping style, and personality factors in the development of breast cancer. Journal of Psychosomatic Research 49, 169–181. Carrera, M.P., Ramirex-Exposito, M.J., Valenzuela, M.T., Garcia, M.J., Mayas, M.D., Martinez-Martos, J.M., 2004. Serum oxytocinase activity is related to tumor growth parameters in N-methyl nitrosourea induced rat breast cancer. Life Sciences 75, 1369–1377. Carter, C.S., 1998. Neuroendocrine perspectives on social attachment and love. Psychoneuroendocrinology 23, 779–818. Cassoni, P., Sapino, A., Marrocco, T., Chini, B., Bussolati, G., 2004. Oxytocin and oxytocin receptors in cancer cells and proliferation. Journal of Neuroendocrinology 16, 362–364. DeVries, A.C., DeVries, M.B., Taymans, S.E., Carter, C.S., 1996. The effects of stress on social preferences are sexually dimorphic in prairie voles. Proceedings of the National Academy of Sciences of the United States of America 93, 11980–11984. Dickerson, S.S., Kemeny, M.E., 2004. Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. Psychological Bulletin 130, 355–391. Kalin, N.H., Gibbs, D.M., Barksdale, C.M., Shelton, S.E., Carnes, M., 1985. Behavioral stress decreases plasma oxytocin concentrations in primates. Life Sciences 36, 1275–1280. Kerr, S.L., Melley, A.M., Travea, L., Pole, M., 2003. The relationship of emotional expression and experience to adult attachment style. Individual Differences Research 1 (2), 108–123. Liberzon, I., Young, E.A., 1997. Effects of stress and glucocorticoids on CNS oxytocin receptor binding. Psychoneuroendocrinology 22, 411–422. Light, K.L., Grewen, K.M., Amico, J.A., 2005. More frequent partner hugs and higher oxytocin levels are linked to lower blood pressure and heart rate in premenopausal women. Biological Psychology 69, 5–21. McCarthy, M.M., 1995. Estrogen modulation of oxytocin and its relation to behavior. Advances in Experimental Medical Biology 395, 235–245. McKenna, M., Zevon, M., Corn, B., Rounds, J., 1999. Psychosocial factors and the development of breast cancer: a meta-analysis. Health Psychology 18, 520–531. Murrell, T.G., 1995. The potential for oxytocin to prevent breast cancer: a hypothesis. Breast Cancer Research and Treatment 35, 225–229. Panksepp, J., Nelson, E., Bekkedal, M., 1997. Brain systems for the mediation of social separation-distress and social-reward. Annals of the New York Academy of Sciences 807, 78–100. Pratt, J.J., 1978. Steroid immunoassay in clinical chemistry. Clinical Chemistry 24, 1869–1890. Salonia, A., Nappi, R.E., Pontillo, M., Daverio, R., Smeraldi, A., Briganti, A., Fabbri, F., Zanni, G., Rigatti, P., Montorsi, F., 2005. Menstrual cycle-related
M. Tops et al. / Biological Psychology 75 (2007) 119–123 changes in plasma oxytocin are relevant to normal sexual function in healthy women. Hormones and Behaviour 47, 164–169. Schelling, G., Kilger, E., Roozendaal, B., de Quervain, D.J.-F., Briegel, J., Dagge, A., Rothenhausler, H.B., Krauseneck, T., Nollert, G., Kapfhammer, H.P., 2004. Stress doses of hydrocortisone, traumatic memories, and symptoms of posttraumatic stress disorder in patients after cardiac surgery: a randomized study. Biological Psychology 55, 627–633. Soravia, L.M., Heinrichs, M., Aerni, A., Maroni, C., Schelling, G., Ehlert, U., Roozendaal, B., de Quervain, D.J.-F., 2006. Glucocorticoids reduce phobic fear in humans. Proceedings of the National Academy of Sciences of the United States of America 103, 5585–5590. Stock, S., Uvna¨s-Moberg, K., 1988. Increased plasma levels of oxytocin in response to afferent electrical stimulation in sciatic and vagal nerves and in response to touch and pinch in anaesthetized rats. Acta Physiologica Scandinavia 132, 29–34. Suslow, T., Donges, U.-S., Kersting, A., Arolt, V., 2000. 20-Item Toronto Alexithymia Scale: do difficulties describing feelings assess proneness to
123
shame instead of difficulties symbolizing emotions? Scandinavian Journal of Psychology 41, 329–334. Tacon, A.M., Caldera, Y.M., Bell, N.J., 2001. Attachment style, emotional control, and breast cancer. Family Systems and Health 19, 319–326. Takagi, S., Ohira, H., 2004. Effects of expression and inhibition of negative emotions on health, mood states, and salivary secretory immunoglobulin A in Japanese mildly depressed undergraduates. Perceptual and Motor Skills 98, 1187–1198. Tops, M., van Peer, J.M., Wester, E.A., Wijers, A.A., Korf, J., 2006. Statedependent regulation of cortical activity by cortisol: an EEG study. Neuroscience Letters 404, 39–43. Uvna¨s-Moberg, K., Bjo¨rkstrand, E., Hillegaart, V., Ahlenius, S., 1999. Oxytocin as a possible mediator of SSRI-induced antidepressant effects. Psychopharmacology 142, 95–101. Uvna¨s-Moberg, K., Arn, I., Magnusson, D., 2005. The psychobiology of emotion: the role of the oxytocinergic system. International Journal of Behavioral Medicine 12, 59–65.