madman
Super Moderator
The metabolic role of prolactin: systematic review, meta-analysis and preclinical considerations (2022)
Giovanni Corona, Giulia Rastrelli, Paolo Comeglio, Federica Guaraldi, Diego Mazzatena, Alessandra Sforza , Linda Vignozzib and Mario Maggi
ABSTRACT
Introduction: Hyperprolactinemia has been proven to induce hypogonadism and metabolic derangements in both genders, while the consequences of prolactin (PRL) deficiency have been poorly investigated.
Areas covered: To systematically review and analyze data from clinical studies focusing on the metabolic consequences of abnormally high prolactin levels (HPRL) and low prolactin levels (LPRL). In addition, data from preclinical studies about underlying pathophysiological mechanisms were summarized and discussed.
Expert opinion: PRL contributes to providing the correct amount of energy to support the mother and the fetus/offspring during pregnancy and lactation, but it also has a homeostatic role. Pathological PRL elevation beyond these physiological conditions, but also its reduction, impairs metabolism and body composition in both genders, increasing the risk of diabetes and cardiovascular events. Hence, hypoprolactinemia should be avoided as much as possible during treatment with dopamine agonists for prolactinomas. Patients with hypoprolactinemia, because of endogenous or iatrogenic conditions, deserve, as those with hyperprolactinemia, careful metabolic assessment.
1. Introduction
Due to apes walking upright and to the relative increase in human cranial dimension, as compared to other mammals, the human female birth canal is relatively narrow, forcing women to have developmentally premature offspring [1,2]. Hence, long-term lactation and maternal care of the newborn are two important aspects of human species perpetuation. Several hormones facilitate these attitudes and behaviors, but the most important is prolactin (PRL) and its receptor. The PRL receptor (PRLR) is a single-pass transmembrane receptor belonging to the cytokine receptor superfamily acting through Janus Kinase (JAK) and Signal Transducer and Activator of Transcription 5 (STAT5). Interestingly, PRLR showed a burst of change in the structure during primate evolution along with its ligand, PRL [3], most probably to adapt to the aforementioned scenario. PRL is a pleiotropic 199 amino acid polypeptide discovered in the early thirties of the last century and produced by many cells throughout the human body but mainly secreted in the bloodstream from the anterior pituitary [4]. It serves many biological functions, but its main role in mammals is to favor milk production by controlling mammary gland development (mammogenesis), the onset of lactation (lactogenesis), and galactopoiesis [4].
In humans [5] and chimpanzees [6] PRL levels are almost twice as high in females as in males, and increases by 10- to 20-fold during pregnancy and lactation, substantiating the important role of this hormone in maternal preparation for childbearing. Unlike other pituitary hormones or hormones from other endocrine glands, a clinical condition characterized by an isolated deficiency of PRL has been scarcely investigated. Recently, three cases of isolated PRL deficiency have been described in female subjects from one family with postpartum alactogenesis, due to a PRL gene mutation [7]. No other phenotype was apparent and fertility, along with normal menstrual cycling, was preserved [7]. The latter finding further corroborates the essential role of PRL in milk production
In contrast to the female gender, the biological function(s) of PRL in the male are less defined and a matter of debate. In several animal models, a trophic effect of PRL on the growth and function of male accessory glands has been demonstrated [8,9]. Accordingly, normal circulating PRL levels in men are associated with a trophic effect on seminal vesicle volume and on its emptying activity, as well as with the amount of ejaculated volume [10]
In men, as well as in women, an abnormal increase in PRL levels (HPRL) is associated with hypogonadotropic hypogonadism due to a hyperprolactinemia-induced reduction of gonadotropin-releasing hormone (GnRH) pulsatility [11]. In addition, in women, there is a clear phenotype associated with an abnormal PRL elevation (Chiari Frommel syndrome: amenorrhea/galactorrhea) that was formerly described in the early 1950s[12]. In contrast, in men, besides the hypogonadism-related symptoms and signs, there is no hyperprolactinemia-associated phenotype, and the only, often reported, symptom is severely reduced sexual desire [13,14]. More than ten years ago we originally described, in a large cohort of men consulting for sexual dysfunction, a syndromic condition we termed ‘hypoprolactinemia’ (LPRL) [15]. The condition was characterized by the association between low prolactin levels with particular psychological, sexual, and metabolic issues, including anxiety, premature ejaculation, increased glucose levels, and diabetes [15]. However, considering the metabolic effects of PRL, there are conflicting results – obtained in both preclinical and clinical studies – showing that either low or high PRL might be associated with consistent metabolic derangements (see for review in [9,16,17]).
*The aim of the present review is to describe the metabolic effects of PRL and its receptor in conditions where PRL levels were abnormally decreased or increased. We will overview, through meta-analysis, clinical results obtained in cross-sectional, longitudinal, and intervention studies, focusing on studies involving either ‘abnormally’ high PRL levels or ‘abnormally’ low PRL levels. Preclinical studies from our and other laboratories will aid in the interpretation of clinical results.
3. Results
3.1. Metabolic consequences of (abnormal) PRL increase
3.1.1. Clinical evidence
3.1.1.1. Observational studies
3.1.1.2. Interventional studies
3.1.2. Preclinical evidence
3.1.2.1. PRLR and adipogenesis
3.1.2.2. PRLR and food intake
3.1.2.3. PRLR and insulin secretion
3.2. Metabolic consequences of (abnormal) PRL decrease
3.2.1. Clinical studies
3.2.2. Preclinical evidence
4. Conclusions
Data derived from clinical evidence as detected by a meta-analytic approach support the role of PRL as a metabolic hormone involved in supporting and storing the required substances to favor mammogenesis, lactogenesis, and galactopoiesis during pregnancy and breastfeeding. Pre-clinical data further corroborate the latter findings. However, it is important to emphasize that several limitations should be recognized and the data derived from the meta-analysis, here reported, should be interpreted with caution. First of all, all the meta-analyzed data were obtained from observational studies, which present an important risk of bias due to the lack of completeness of follow-up and the accrual of missing data [79]. In addition, specific sub-analyses limited to only male or female populations were available only in a limited number of studies. Significant heterogeneity among studies was detected, which reflects the differences observed in population characteristics and in the type of DA preparations and dosages used. It is well known that levels of PRL have a high variability because of its pulsatile release and due to its regulation by a large number of physiological factors, including estrogen levels. Additionally, PRL concentrations tend to be altered based on the phase of the menstrual cycle, and contraceptive use. Unfortunately, information on all these factors was available only in a limited number of the studies included and no further analyses were possible.
The concept of LPRL as a clinical entity was introduced quite recently [32] and only a few studies have investigated this condition either in males or females. The criteria used for the definition of both HPRL and LPRL differ among the studies. The characteristics of control groups differ among studies. Subjects in the HPRL group tend to be younger than those included in the LPRL group, which represents a further source of bias. The magnitude of observed differences between HPRL/LPRL and controls derived from our meta-analysis is quite small, suggesting that other factors may well play a possible role. Finally, no information on the effect of increasing PRL in patients with LPRL is available. Hence, the reproducibility of our data warrants caution.
5. Expert opinion
The clinical studies summarized here essentially indicate that PRL is not only a reproductive but also a metabolic hormone. This is likely because the two functions are intimately interconnected, also considering that reproductive function is a costly process in terms of energy consumption. Pregnancy and lactation are clear examples and in these particular conditions, PRL and its receptor play an important role: in one way storing and in the other one delivering nutrients to the fetus and the newborn. Accordingly, preclinical studies reviewed elsewhere [9,16,17], and the results presented here indicate that PRLR is expressed in tissues regulating not only food intake (hypothalamus) and fat handling (adipose tissue) but also insulin secretion, such as in the pancreas. Hence, it is not surprising that in conditions that mimic the pregnancy-induced PRL increase – i.e. any pathological hyperprolactinemic state – there is increased body weight, increased waist circumference, and fat accumulation, along with insulin resistance. The increase in fat mass associated with prolactinoma might also be due to the concomitant HPRL-induced hypogonadotropic hypogonadism, at least in males. In fact, in males, T deficiency is associated with an increase in fat mass [80]. Treating prolactinoma is associated with a reduction of BMI, a reduction in glycemia, and the amelioration of dyslipidemia, along with a reduction of insulin resistance. It is interesting to note that meta-regression analysis of clinical studies indicates that the positive effects of prolactinoma therapy on fasting glucose and lipids are more apparent in females than in males, whereas the opposite trend was observed for BMI. These observations are in line with what was reported in pre-clinical models [54]. Recent data indicate that DAT is able to restore normal T levels in no more than 2/3 of patients with macroprolactinomas [81]. Available guidelines indicate adding TRT to DAT when the latter therapy alone is not able to completely restore normal T levels [82,83]. Data from the general population [80,84,85] as well as from patients with T2DM or MetS [22,86] have shown that TRT can clearly modify body composition, by reducing fat mass and improving lean mass, however, its role on lipid profile and glycometabolic control is more conflicting [87–89]. Hence, the more limited effects on fasting glycemia and total cholesterol observed in males in the present study after DAT can be explained, at least partially, by the persistence of reduced T levels at the end-point. Unfortunately, the latter information was available only in one study [39], included in our analysis, which confirmed mean reduced T levels at follow-up.
The hyperprolactinemia-induced increase in circulating lipid and glucose does not lead to a state of overt diabetes, because insulin secretion is also increased, most probably due to a PRL-induced stimulation in β-cell secretory response and β-cell mass. The latter may overcome the state of insulin resistance, as usually observed in normal pregnancy. Accordingly, low maternal PRL during pregnancy predicts postpartum prediabetes/diabetes [90], most probably because the compensatory stimulation of insulin secretion is insufficient. In line with this evidence, the present meta-analysis shows that low PRL is associated with overt diabetes in cross-sectional studies and with the risk of developing diabetes in longitudinal studies.
Soto-Pedre et al. [91], reported that high PRL due to pituitary microadenomas was not associated with increased overall mortality, whereas a higher mortality risk was observed in patients with macroadenomas and in those with drug-induced and idiopathic hyperprolactinemia. Other studies supported high CV risk related to HPRL in males but not in females [92,93]. In particular, a recent retrospective observational study including a total of 3,633 patients with a median follow-up time of 5.3 years showed that hyperprolactinemia was associated with higher CV mortality and morbidity risk in males but not in females [93]. The same study also documented that the adjustment for the use of antipsychotic medication attenuated the observed risk [93]. The specific underlying mechanisms supporting the latter gender difference have yet to be better elucidated. The higher risk observed in patients with macro-adenomas, which are usually characterized by higher PRL circulating levels, suggests a possible role of reduced T levels in the stratification of HPRL-induced CV risk [94]. Conversely, the association with the use of antipsychotic medications points out other possibilities revised elsewhere [95,96]. Conversely to what was observed for HPRL, low PRL was associated with increased major adverse CV events in high-risk subjects [23] and with a higher incidence of left ventricular altered geometry and hypertrophy during five years of follow-up in the SHIP (Study of Health in Pomerania) population-based study [97].
*Based on the available data, it is our expert opinion that the real problem for PRL-associated metabolic derangements is a decreased and not an increased circulating PRL. In fact, physiological hyperprolactinemia, as observed during pregnancy and lactation, has a homeostatic significance, allowing for correct energy distribution between the mother and the fetus/offspring. In other words, PRL is not a diabetogenic hormone but actually shows anti-diabetogenic effects. Accordingly, its deficiency is associated with an increased risk of diabetes and cardiovascular events. The clinical syndrome hypoprolactinemia is a puzzling, new condition that needs further studies to define its pathological burden. Further studies are advisable to better clarify our hypothesis.
Giovanni Corona, Giulia Rastrelli, Paolo Comeglio, Federica Guaraldi, Diego Mazzatena, Alessandra Sforza , Linda Vignozzib and Mario Maggi
ABSTRACT
Introduction: Hyperprolactinemia has been proven to induce hypogonadism and metabolic derangements in both genders, while the consequences of prolactin (PRL) deficiency have been poorly investigated.
Areas covered: To systematically review and analyze data from clinical studies focusing on the metabolic consequences of abnormally high prolactin levels (HPRL) and low prolactin levels (LPRL). In addition, data from preclinical studies about underlying pathophysiological mechanisms were summarized and discussed.
Expert opinion: PRL contributes to providing the correct amount of energy to support the mother and the fetus/offspring during pregnancy and lactation, but it also has a homeostatic role. Pathological PRL elevation beyond these physiological conditions, but also its reduction, impairs metabolism and body composition in both genders, increasing the risk of diabetes and cardiovascular events. Hence, hypoprolactinemia should be avoided as much as possible during treatment with dopamine agonists for prolactinomas. Patients with hypoprolactinemia, because of endogenous or iatrogenic conditions, deserve, as those with hyperprolactinemia, careful metabolic assessment.
1. Introduction
Due to apes walking upright and to the relative increase in human cranial dimension, as compared to other mammals, the human female birth canal is relatively narrow, forcing women to have developmentally premature offspring [1,2]. Hence, long-term lactation and maternal care of the newborn are two important aspects of human species perpetuation. Several hormones facilitate these attitudes and behaviors, but the most important is prolactin (PRL) and its receptor. The PRL receptor (PRLR) is a single-pass transmembrane receptor belonging to the cytokine receptor superfamily acting through Janus Kinase (JAK) and Signal Transducer and Activator of Transcription 5 (STAT5). Interestingly, PRLR showed a burst of change in the structure during primate evolution along with its ligand, PRL [3], most probably to adapt to the aforementioned scenario. PRL is a pleiotropic 199 amino acid polypeptide discovered in the early thirties of the last century and produced by many cells throughout the human body but mainly secreted in the bloodstream from the anterior pituitary [4]. It serves many biological functions, but its main role in mammals is to favor milk production by controlling mammary gland development (mammogenesis), the onset of lactation (lactogenesis), and galactopoiesis [4].
In humans [5] and chimpanzees [6] PRL levels are almost twice as high in females as in males, and increases by 10- to 20-fold during pregnancy and lactation, substantiating the important role of this hormone in maternal preparation for childbearing. Unlike other pituitary hormones or hormones from other endocrine glands, a clinical condition characterized by an isolated deficiency of PRL has been scarcely investigated. Recently, three cases of isolated PRL deficiency have been described in female subjects from one family with postpartum alactogenesis, due to a PRL gene mutation [7]. No other phenotype was apparent and fertility, along with normal menstrual cycling, was preserved [7]. The latter finding further corroborates the essential role of PRL in milk production
In contrast to the female gender, the biological function(s) of PRL in the male are less defined and a matter of debate. In several animal models, a trophic effect of PRL on the growth and function of male accessory glands has been demonstrated [8,9]. Accordingly, normal circulating PRL levels in men are associated with a trophic effect on seminal vesicle volume and on its emptying activity, as well as with the amount of ejaculated volume [10]
In men, as well as in women, an abnormal increase in PRL levels (HPRL) is associated with hypogonadotropic hypogonadism due to a hyperprolactinemia-induced reduction of gonadotropin-releasing hormone (GnRH) pulsatility [11]. In addition, in women, there is a clear phenotype associated with an abnormal PRL elevation (Chiari Frommel syndrome: amenorrhea/galactorrhea) that was formerly described in the early 1950s[12]. In contrast, in men, besides the hypogonadism-related symptoms and signs, there is no hyperprolactinemia-associated phenotype, and the only, often reported, symptom is severely reduced sexual desire [13,14]. More than ten years ago we originally described, in a large cohort of men consulting for sexual dysfunction, a syndromic condition we termed ‘hypoprolactinemia’ (LPRL) [15]. The condition was characterized by the association between low prolactin levels with particular psychological, sexual, and metabolic issues, including anxiety, premature ejaculation, increased glucose levels, and diabetes [15]. However, considering the metabolic effects of PRL, there are conflicting results – obtained in both preclinical and clinical studies – showing that either low or high PRL might be associated with consistent metabolic derangements (see for review in [9,16,17]).
*The aim of the present review is to describe the metabolic effects of PRL and its receptor in conditions where PRL levels were abnormally decreased or increased. We will overview, through meta-analysis, clinical results obtained in cross-sectional, longitudinal, and intervention studies, focusing on studies involving either ‘abnormally’ high PRL levels or ‘abnormally’ low PRL levels. Preclinical studies from our and other laboratories will aid in the interpretation of clinical results.
3. Results
3.1. Metabolic consequences of (abnormal) PRL increase
3.1.1. Clinical evidence
3.1.1.1. Observational studies
3.1.1.2. Interventional studies
3.1.2. Preclinical evidence
3.1.2.1. PRLR and adipogenesis
3.1.2.2. PRLR and food intake
3.1.2.3. PRLR and insulin secretion
3.2. Metabolic consequences of (abnormal) PRL decrease
3.2.1. Clinical studies
3.2.2. Preclinical evidence
4. Conclusions
Data derived from clinical evidence as detected by a meta-analytic approach support the role of PRL as a metabolic hormone involved in supporting and storing the required substances to favor mammogenesis, lactogenesis, and galactopoiesis during pregnancy and breastfeeding. Pre-clinical data further corroborate the latter findings. However, it is important to emphasize that several limitations should be recognized and the data derived from the meta-analysis, here reported, should be interpreted with caution. First of all, all the meta-analyzed data were obtained from observational studies, which present an important risk of bias due to the lack of completeness of follow-up and the accrual of missing data [79]. In addition, specific sub-analyses limited to only male or female populations were available only in a limited number of studies. Significant heterogeneity among studies was detected, which reflects the differences observed in population characteristics and in the type of DA preparations and dosages used. It is well known that levels of PRL have a high variability because of its pulsatile release and due to its regulation by a large number of physiological factors, including estrogen levels. Additionally, PRL concentrations tend to be altered based on the phase of the menstrual cycle, and contraceptive use. Unfortunately, information on all these factors was available only in a limited number of the studies included and no further analyses were possible.
The concept of LPRL as a clinical entity was introduced quite recently [32] and only a few studies have investigated this condition either in males or females. The criteria used for the definition of both HPRL and LPRL differ among the studies. The characteristics of control groups differ among studies. Subjects in the HPRL group tend to be younger than those included in the LPRL group, which represents a further source of bias. The magnitude of observed differences between HPRL/LPRL and controls derived from our meta-analysis is quite small, suggesting that other factors may well play a possible role. Finally, no information on the effect of increasing PRL in patients with LPRL is available. Hence, the reproducibility of our data warrants caution.
5. Expert opinion
The clinical studies summarized here essentially indicate that PRL is not only a reproductive but also a metabolic hormone. This is likely because the two functions are intimately interconnected, also considering that reproductive function is a costly process in terms of energy consumption. Pregnancy and lactation are clear examples and in these particular conditions, PRL and its receptor play an important role: in one way storing and in the other one delivering nutrients to the fetus and the newborn. Accordingly, preclinical studies reviewed elsewhere [9,16,17], and the results presented here indicate that PRLR is expressed in tissues regulating not only food intake (hypothalamus) and fat handling (adipose tissue) but also insulin secretion, such as in the pancreas. Hence, it is not surprising that in conditions that mimic the pregnancy-induced PRL increase – i.e. any pathological hyperprolactinemic state – there is increased body weight, increased waist circumference, and fat accumulation, along with insulin resistance. The increase in fat mass associated with prolactinoma might also be due to the concomitant HPRL-induced hypogonadotropic hypogonadism, at least in males. In fact, in males, T deficiency is associated with an increase in fat mass [80]. Treating prolactinoma is associated with a reduction of BMI, a reduction in glycemia, and the amelioration of dyslipidemia, along with a reduction of insulin resistance. It is interesting to note that meta-regression analysis of clinical studies indicates that the positive effects of prolactinoma therapy on fasting glucose and lipids are more apparent in females than in males, whereas the opposite trend was observed for BMI. These observations are in line with what was reported in pre-clinical models [54]. Recent data indicate that DAT is able to restore normal T levels in no more than 2/3 of patients with macroprolactinomas [81]. Available guidelines indicate adding TRT to DAT when the latter therapy alone is not able to completely restore normal T levels [82,83]. Data from the general population [80,84,85] as well as from patients with T2DM or MetS [22,86] have shown that TRT can clearly modify body composition, by reducing fat mass and improving lean mass, however, its role on lipid profile and glycometabolic control is more conflicting [87–89]. Hence, the more limited effects on fasting glycemia and total cholesterol observed in males in the present study after DAT can be explained, at least partially, by the persistence of reduced T levels at the end-point. Unfortunately, the latter information was available only in one study [39], included in our analysis, which confirmed mean reduced T levels at follow-up.
The hyperprolactinemia-induced increase in circulating lipid and glucose does not lead to a state of overt diabetes, because insulin secretion is also increased, most probably due to a PRL-induced stimulation in β-cell secretory response and β-cell mass. The latter may overcome the state of insulin resistance, as usually observed in normal pregnancy. Accordingly, low maternal PRL during pregnancy predicts postpartum prediabetes/diabetes [90], most probably because the compensatory stimulation of insulin secretion is insufficient. In line with this evidence, the present meta-analysis shows that low PRL is associated with overt diabetes in cross-sectional studies and with the risk of developing diabetes in longitudinal studies.
Soto-Pedre et al. [91], reported that high PRL due to pituitary microadenomas was not associated with increased overall mortality, whereas a higher mortality risk was observed in patients with macroadenomas and in those with drug-induced and idiopathic hyperprolactinemia. Other studies supported high CV risk related to HPRL in males but not in females [92,93]. In particular, a recent retrospective observational study including a total of 3,633 patients with a median follow-up time of 5.3 years showed that hyperprolactinemia was associated with higher CV mortality and morbidity risk in males but not in females [93]. The same study also documented that the adjustment for the use of antipsychotic medication attenuated the observed risk [93]. The specific underlying mechanisms supporting the latter gender difference have yet to be better elucidated. The higher risk observed in patients with macro-adenomas, which are usually characterized by higher PRL circulating levels, suggests a possible role of reduced T levels in the stratification of HPRL-induced CV risk [94]. Conversely, the association with the use of antipsychotic medications points out other possibilities revised elsewhere [95,96]. Conversely to what was observed for HPRL, low PRL was associated with increased major adverse CV events in high-risk subjects [23] and with a higher incidence of left ventricular altered geometry and hypertrophy during five years of follow-up in the SHIP (Study of Health in Pomerania) population-based study [97].
*Based on the available data, it is our expert opinion that the real problem for PRL-associated metabolic derangements is a decreased and not an increased circulating PRL. In fact, physiological hyperprolactinemia, as observed during pregnancy and lactation, has a homeostatic significance, allowing for correct energy distribution between the mother and the fetus/offspring. In other words, PRL is not a diabetogenic hormone but actually shows anti-diabetogenic effects. Accordingly, its deficiency is associated with an increased risk of diabetes and cardiovascular events. The clinical syndrome hypoprolactinemia is a puzzling, new condition that needs further studies to define its pathological burden. Further studies are advisable to better clarify our hypothesis.
