Religion paper

Volume 76, Part B, January 2014, Pages 269–275

NIDA 40th Anniversary Issue

125461||

Invited review

The impact of exposure to addictive drugs on future generations: Physiological and behavioral effects

· F.M. Vassoler a , , , 

· E.M. Byrnes a , 

· R.C. Pierce b

 Show more

doi:10.1016/j.neuropharm.2013.06.016

Get rights and content

Highlights

•

Transgenerational consequences of exposure to drugs of abuse.

•

Paternal and maternal transmission to offspring prior to conception.

•

Alcohol,  nicotine ,  cannabinoids ,  opioids ,  cocaine .

Abstract

It is clear that both genetic and environmental factors contribute to drug addiction. Recent evidence indicating trans-generational influences of drug abuse highlight potential  epigenetic  factors as well. Specifically, mounting evidence suggests that parental ingestion of abused drugs influence the physiology and behavior of future generations even in the absence of  prenatal  exposure. The goal of this review is to describe the trans-generational consequences of preconception exposure to drugs of abuse for five major classes of drugs: alcohol,  nicotine , marijuana,  opioids , and  cocaine . The potential epigenetic mechanisms underlying the transmission of these  phenotypes across generations also are detailed.

This article is part of a Special Issue entitled ‘NIDA 40th Anniversary Issue’.

Keywords

· Transgenerational; 

· Epigenetic ; 

· Drugs of abuse; 

· Alcohol; 

· Nicotine ; 

· Marijuana; 

· Opioids ;

· Cocaine

1. Introduction

Drug addiction is a serious medical and social issue in the United States and around the world. There is consistent evidence that  substance use disorders  run in families ( Bierut et al., 1998 ,  Brook et al., 2002 ,  Cloninger et al., 1981  and  Merikangas et al., 1998 ). Adoption, twin, and sibling studies implicate genetic factors in the heritability of abuse ( Cloninger et al., 1981 ). However, simple genetic mechanisms of inheritance cannot explain all results ( Cloninger et al., 1981  and  Schuckit et al., 1972 ). Societal differences in drug use and consumption patterns vary from different time periods and between countries suggesting a large environmental component ( UNODC, 2012 ). Thus, vulnerability to develop an addiction is dependent on both genetics and the environment. The sum of separate genetic and environmental contributions cannot fully explain the heritability either. Therefore, the interaction between genetics and the environment may help explain some of the discrepancies ( Cloninger et al., 1981 ).

Epigenetics  is a key mechanism by which the environment can influence and interact with genetics. In recent years, the term epigenetics has been used to describe myriad processes ( Haig, 2004 ). For example, modifications to the structure of  chromatin  or DNA without changes in the sequence that affect gene transcription even in non-dividing cells such as  DNA methylation  or  histone acetylation  are described as  epigenetic modifications ( Holliday, 1989 ). However, some definitions of epigenetics emphasize that the modifications in gene expression that do not involve alterations in the  DNA sequence must be heritable, spanning multiple generations. There has recently been an increase in studies examining the transgenerational effects of environmental toxins on offspring. The goal of the current review is to examine the available literature regarding the effects on offspring of parental drug exposure in the absence of any direct fetal exposure for five major drugs of abuse. While offspring susceptibility to drug use is of particular interest, all behavioral, molecular, and physiological changes in the offspring will be reported. The majority of the studies available on this topic focus on paternal transmission of epigenetic phenotypes , as this model eliminates any direct fetal exposure and ostensibly avoids maternal rearing effects. With regard to maternal transmission, there is certainly an extensive body of literature documenting offspring effects following  prenatal  exposure to drugs of abuse ( Malanga and Kosofsky, 2003  and  Sithisarn et al., 2012 ). Due to the possible direct effects of in utero exposure on the fetus, as well as the numerous confounds that drug use during pregnancy introduces (e.g. changes in nutritional status, renal function, vascular perfusion, etc.), prenatal substance use models will not be included in the current review. We will, however, include data from studies examining transgenerational effects of female drug use occurring prior to conception. Thus, the current review will examine the effects of parental exposure to drugs of abuse prior to conception on the development of subsequent generations.

2. Alcohol

Currently, alcohol is the most commonly abused drug in the United States. In 2011, the center for disease control estimated that 60% of males and 44% of females engage in chronic alcohol drinking ( Edward J. Sondik et al., 2012 ). While the neurobehavioral effects of fetal alcohol exposure are well described, less is known about the effects of parental exposure to alcohol prior to conception. It should be noted, however, that reports and writings as early as the 1720's, during the so-called gin epidemic, observed that both maternal and paternal alcohol use had detrimental effects on offspring ( Warner and Rosett, 1975 ). While few studies have examined maternal alcohol use prior to pregnancy (i.e. in the absence of  prenatal  use), several findings have demonstrated effects of paternal alcohol exposure on offspring development. Indeed, as early as 1913 animal studies found that offspring sired by alcohol inhaling rats demonstrated malformations, low birth weight, retarded growth and increased neonatal mortality across several generations ( Friedler, 1996 ). Clearly the concept that alcohol use by the father, and not solely his genetic composition, can affect future progeny is not novel. The interest in paternal effects, however, has been reinvigorated by the emergence of the field of  epigenetics , with a number of preclinical findings suggesting transgenerational epigenetic effects  of paternal alcohol use.

Several of the initial animal studies on paternal alcohol effects focused on basic parameters of reproductive success, such as fertility and fecundity, following exposure to alcohol in peripubertal males. These studies observed a significant reduction in the number of successful pregnancies which decreased from 92% in controls to 75% in naïve females mated with alcohol-drinking sires ( Emanuele et al., 2001 ). Litter size was also substantially reduced ( Cicero et al., 1990  and  Emanuele et al., 2001 ). It was determined that alcohol exposure during puberty modified sexual maturation and resulted in decreased testes and secondary sex organ weight, eliminated the typical pubertal surge in  testosterone , decreased  beta-endorphin  levels in the  hypothalamus , and enhanced testicular oxidative injury ( Cicero et al., 1990  and  Emanuele et al., 2001 ). Interestingly, offspring of these pubertal alcohol-exposed sires demonstrated similar alterations, including decreased serum testosterone levels, reduced  seminal vesicle weights, and lower levels of  hypothalamic  beta-endorphin ( Abel and Lee, 1988  and  Cicero et al., 1990 ).

Alcohol-sired offspring also demonstrate abnormalities in development. In an elegantly designed rodent study, Jamerson and colleagues revealed that paternal alcohol exposure that was ongoing, or that had ceased weeks prior to conception resulted in more rapid development of various reflexes, differences in  gait , and thicker  cortical layers ( Jamerson et al., 2004 ). It was also noted that timing of alcohol exposure in relation to conception impacted the neurobehavioral effects of the offspring ( Jamerson et al., 2004 ). For example, one study found that a single exposure to alcohol just prior to conception resulted in a significant increase in small for gestational age offspring as well as an increase in offspring demonstrating significant malformations ( Bielawski and Abel, 1997 ). Moreover, alcohol-exposed sires also produced offspring displaying increased adrenal cortex  and decreased  spleen  weights ( Abel, 1993b ). Finally, there is evidence that metabolic and immune functioning may be disrupted in alcohol-sired offspring, given that they display reductions in  leptin  levels ( Emanuele et al., 2001 ) and a diminished immune response ( Berk et al., 1989  and  Hazlett et al., 1989 ).

In terms of behavioral effects, both increases and decreases in activity have been noted in offspring of alcohol-consuming sires ( Abel, 1989a ,  Abel, 1989b ,  Abel, 1993a ,  Abel, 1993b  and  Abel and Lee, 1988 ), with the direction of these effects mediated by a number of factors including the level of alcohol consumption, the time between exposure and conception ( Jamerson et al., 2004 ), and the age at the time of testing ( Abel, 1989a ). Alterations in behavioral activity following  amphetamine  were also noted in male offspring ( Abel, 1993a ), with some evidence suggesting that increased activity was dependent on the  cholinergic system  ( Abel, 1994 ). Potential modifications in the cholinergic system of alcohol-sired offspring are notable given that deficits in learning and memory have also been reported. For example, offspring sired by alcohol treated males demonstrated impairments in  spatial learning  ( Wozniak et al., 1991 ) and had increased latencies to reach a choice point in a  T-maze  ( Abel, 1994  and  Abel and Lee, 1988 ). In addition to deficits observed in males, alcohol-sired female offspring showed impaired performance in a two-way shock  avoidance learning  task ( Abel and Tan, 1988 ). Finally, offspring of alcohol-consuming sires demonstrated decreased grooming as well as decreased immobility in a  forced swim test , an effect that was rescued by  imipramine and  propranolol  and exacerbated by  yohimbine  and  metergoline  ( Abel, 1991a ,  Abel, 1991b  and  Abel and Bilitzke, 1990 ). Together, these results indicate a detrimental behavioral  phenotype  of paternal alcohol consumption on both male and female offspring.

Examining  epigenetic  parental effects in human populations can be difficult. Human studies have, however, revealed correlations between parental drinking behavior and offspring initiation and drinking patterns. Thus, heavy paternal drinking or heavy-episodic drinking in both parents predicts earlier onset of offspring drinking as well as heavier drinking ( Vermeulen-Smit et al., 2012 ). Children born with characteristics associated with fetal alcohol syndrome whose mother's did not drink but fathers were alcoholics have also been observed ( Lemoine et al., 2003 ). Additionally, there is evidence that sons of early-onset (adolescent) alcoholic fathers perform more poorly on tests of verbal intelligence  and attention than late onset (adult) alcoholics ( Tarter et al., 1989 ). While this may suggest that adolescence, or periods of significant neural and  endocrine development, are particularly important in determining the effects of paternal alcohol exposure, one cannot exclude the possibility that early-onset drinking may be a marker of genetic vulnerability or additional  comorbid  pathologies. Finally, in addition to deficits in cognition, attention, and visuospatial capacity observed in children of alcoholic fathers, increased hyperactivity has also been noted ( Goodwin et al., 1975 ), although the relationship between hyperactivity and paternal alcohol consumption remains equivocal ( Knopik et al., 2009 ). Much less work has been done examining the effects of maternal preconception alcohol consumption. However, a decreased birth weight has been observed in children of alcoholic women that abstain from use during pregnancy ( Little et al., 1980 ,  Livy et al., 2004  and  Ramsay, 2010 ). While human studies remain more difficult to interpret than preclinical research due to many aspects that are impossible to control, it is important that both lines of research continue in order to fully understand the scope of preconception alcohol use and abuse.

3. Nicotine

Nicotine  is the second most abused substance in the United States. In 2011 it was reported that 21.6% of adult men and 16.5% of adult women smoke regularly ( Agaku et al., 2011 ). It is clear from both human epidemiological and preclinical research models that exposure to tobacco smoke in utero is harmful to development and can result in low birth weight, sudden infant death syndrome, and serious behavioral issues in the offspring (  Abbott and Winzer-Serhan, 2012 ). Examining the indirect effects (i.e. non- prenatal  exposure) in humans is more challenging because it is difficult to distinguish the indirect  epigenetic effects  of paternal or maternal nicotine abuse on the offspring from potential exposure to second-hand smoke in utero or post-partum. Keeping that in mind, some studies have shown increased risk of spontaneous abortion associated with paternal smoking (  Blanco-Munoz et al., 2009  and  Venners et al., 2004 ). However, others have shown no association (  Chatenoud et al., 1998 ,  Windham et al., 1992  and  Windham et al., 1999 ). Moreover, paternal smoking is a factor associated with anorectal malformations in humans in multiple studies ( Zwink et al., 2011 ). Human epidemiological studies also indicate that early paternal smoking (onset before age 11) is associated with greater  body mass index  in male, but not female, offspring (  Pembrey, 2010  and  Pembrey et al., 2006 ).

Aside from these few studies looking at the effect of paternal smoking on offspring, numerous studies have found evidence that male smoking affects multiple fertility factors in the male. Thus, paternal tobacco use is associated with defects in the tail of the spermatozoon ( Ozgur et al., 2005 ) and decreases in sperm density, motility, and morphology ( Sofikitis et al., 1995 ). Another study found no differences in sperm concentration or motility but decreased semen volume ( Pasqualotto et al., 2006 ). Taken together, these studies point toward decreases in male fertility and/or embryo viability which suggests potential negative effects in the offspring. To date, we found only one animal study which found that paternal nicotine smoke exposure did not affect development or any behavioral parameters measured ( Gaworski et al., 2004 ). More animal work is needed to determine if nicotine exposure and smoking has transgenerational epigenetic effects.

4. Cannabinoids

Marijuana is the most widely abused illicit substance in the United States. Of particular concern is the high percentage of adolescents that engage in marijuana use. It is estimated that almost 50% of teenagers have used marijuana and approximately 9% use marijuana heavily (>20 days out of the month) ( Metlife, 2011 ). Marijuana use by adolescents may be particularly problematic as developing systems may be more vulnerable to the impact of exogenous  cannabinoids  ( Rice and Barone, 2000 ). Abuse by adolescent females is of particular concern as the  endogenous cannabinoid system  is important for reproductive physiology, with  CB1 receptors  as well as  endogenous cannabinoids  expressed in the ovaries, uterine endometrium, and other peripheral endocrine  tissue ( Bari et al., 2011 ). Thus,  cannabinoid  exposure during this critical period could result in lasting modifications in female reproduction, and might impact future offspring. To examine the potential effects on offspring, adolescent females were exposed to cannabinoids during adolescence. They were then maintained drug-free for over 3 weeks and bred during adulthood. Using this model, it was shown that both adolescent and adult male offspring of adolescently-exposed dams exhibit greater sensitivity to  morphine -induced  conditioned place preference  than the control animals, even in the absence of any direct in utero exposure (  Byrnes et al., 2012 ). Moreover, unpublished data from the Byrnes laboratory shows that female offspring demonstrate enhanced expression of morphine-induced locomotor sensitization accompanied by increased expression of the  mu opioid receptor  in the  nucleus accumbens .

While no studies to date have examined the effect on the offspring of cannabinoid exposure in the sire prior to conception, available evidence suggests that disruptions in endocrine functioning resulting from cannabinoid exposure may cause deleterious effects in subsequent generations. For example,  tetrahydrocannabinol  ( THC ) disrupts gonadal functions by depriving testicular cells of energy reserves and stimulating androgen-binding protein  secretion which leads to oligospermia in chronic  cannabis smokers. THC also interferes with production of  prolactin ,  luteinizing hormone  ( LH ), and follicle stimulating hormone  ( FSH ), which causes reduced  testosterone  production in the testes ( Banerjee et al., 2011 ,  Harclerode, 1984  and  Husain and Khan, 1985 ). It has also been suggested that acute cannabinoid treatment affects the quality and quantity of spermatozoa produced by the testis ( Harclerode, 1984 ). Animal models have demonstrated that cannabinoid administration suppresses  gonadal steroids ,  growth hormone , prolactin,  thyroid hormone  and activates the  HPA-axis  ( Banerjee et al., 2011  and  Brown and Dobs, 2002 ). These effects have been shown to be mediated by the binding of exogenous cannabinoids to the  endogenous  receptor in or near the hypothalamus  ( Brown and Dobs, 2002 ). However, the effects in humans have been inconsistent likely due to the development of tolerance ( Brown and Dobs, 2002 ). Finally, in humans, two case controlled studies found an increased risk for congenital heart defects associated with paternal marijuana use ( Ewing et al., 1997  and  Wilson et al., 1998 ). Clearly, there is a large gap in the literature examining the effects of marijuana use preconception on future generations.

5. Morphine

The significant impact of maternal  opioid  use on fetal outcomes and offspring neurodevelopment has been an area of interest for decades ( Hutchings, 1982 ,  Malanga and Kosofsky, 1999  and  Vathy, 2002 ). In both substance abusing women and animal models of  prenatal  substance use the fetus is directly exposed to  opioids , limiting the applicability to a discussion on transgenerational  epigenetic  processes. To avoid direct fetal exposure, animal models were developed that expose both males and females to opioids preconception and then examine the effects on their offspring. Such studies have primarily examined the impact of exposure to  morphine  during adolescence. In male rats, morphine administered throughout adolescent development resulted in significant modifications in sexual maturation in the males themselves. When these males were mated to  drug naïve  females several weeks after morphine exposure, they produced smaller litters and as adults their offspring demonstrated significant modifications in endocrine  parameters, including sex specific changes in adrenal weights,  luteinizing hormone  and  hypothalamic  β-endorphin ( Cicero et al., 1991 ). As all that the male contributed in this model was sperm, such findings strongly suggest that prior exposure to opioids can induce transgenerational  epigenetic effects  even in the absence of continuing use.

Similar findings have been revealed using a female rat model. In those studies, females were exposed to an intermittent, increasing dose regimen of morphine during early-mid adolescence. All animals were then drug-free for several weeks prior to mating with drug naïve males. In these studies there were no significant differences in any postnatal parameter (i.e. litter size, weight, sex ratio). When offspring were examined as adults, however, a number of significant effects have been observed. These include sex-specific changes in social and emotional behaviors ( Byrnes et al., 2011  and  Johnson et al., 2011 ) as well as alterations in their response to morphine ( Byrnes, 2005  and  Byrnes et al., 2011 ). More recent findings using this model also observed significant changes in the functional response to  dopamine agonists  coupled with increased levels of dopamine- and opioid-related gene expression in the  nucleus accumbens . Of note, these effects were observed in both the first (F1) and second (F2) generation, suggesting multigenerational epigenetic effects triggered by adolescent exposure ( Byrnes et al., 2013 ).

The mechanisms underlying such intergenerational transmission are unknown and perhaps more complicated when considering maternal transmission. Exposure to drugs of abuse, even when experienced prior to conception, may alter the prenatal hormonal milieu, thereby modifying the developmental trajectory of offspring. Alternatively, or perhaps additionally, modifications in maternal care as a result of prior exposure to opioids, could impact the development of offspring. Indeed, morphine administered both prior to parturition or during active mothering significantly alters maternal care ( Bridges and Grimm, 1982 ,  Mann et al., 1991 ,  Miranda-Paiva et al., 2001  and  Slamberova et al., 2001 ). Moreover, non-genomic transmission via subtle variations in maternal care have been demonstrated repeatedly in animal models ( McLeod et al., 2007 ,  Szyf et al., 2007 , Weaver, 2007  and  Weaver et al., 2004 ). Thus, one possible mechanism of transmission may be alterations in the pre- and/or postnatal environment.

Alternatively, exposure to opioids could directly affect epigenetic  germline cells  in the exposed male and female. Such effects could alter early embryogenesis and impact a number of developmental processes. These effects could then be passed forward via alterations in behavior/physiology of the next generation or via direct  epigenetic inheritance . Additional studies are needed to determine the mechanisms underlying transmission in both males and females exposed to opioids. Nonetheless, the nature of the observed offspring effects, including dysregulation of the  hypothalamic–pituitary–adrenal axis , increased sensitivity to opioids, and blunted response to dopamine agonists, all suggest that parental exposure to opioids prior to conception may increase the risk of substance use in their future offspring.

6. Cocaine

In terms of parental  cocaine  administration, the clinical and preclinical literatures have focused primarily on the developmental effects of  prenatal  cocaine exposure. In that regard, preclinical studies relatively consistently show disrupted cortical and hippocampal  development in the offspring as well as impaired  executive function  and memory following in utero cocaine exposure (  Dow-Edwards, 2011 ,  Lidow, 2003  and  Malanga and Kosofsky, 2003 ).  Epigenetic  mechanisms appear to play a role in these changes in that male mice exposed to cocaine prenatally had altered patterns of DNA methylation  in hippocampal  pyramidal neurons  ( Novikova et al., 2008 ), which could underlie impaired sustained attention and  spatial working memory  observed in these offspring ( He et al., 2006b ). While interesting, studies of this sort are directly examining the effect of cocaine on development. Moreover, the dams were exposed to cocaine, which may have influenced maternal behaviors. These issues are largely obviated in experiments in which paternal cocaine exposure is examined.

It has long been known that cocaine concentrates in the testes at levels second only to the brain ( Mule et al., 1977  and  Yazigi et al., 1991 ) due primarily to specific binding sites in spermatozoa ( Yazigi et al., 1991 ). Animal studies showed that prolonged experimenter-administered cocaine (72–150 days) impaired spermatogenesis ( George et al., 1996 ) and increased the percentage of sperm with tails separated from their heads ( Abel et al., 1989 ). Although some evidence indicated that the fertility of cocaine-treated rats was substantially impaired ( George et al., 1996 ), fecundity remained sufficient to examine the effects of paternal cocaine exposure on their offspring. For example, paternal cocaine exposure decreased the weight of their progeny ( George et al., 1996  and  Killinger et al., 2012 ) but see also ( Abel et al., 1989 ). Experimenter-delivered cocaine to mouse sires also resulted in increased immobility in the  tail suspension test , a model of depression, but had no effect on locomotor activity, measures of anxiety or learning and memory ( Killinger et al., 2012 ). In contrast, the offspring of mouse sires that self-administered cocaine displayed attention and spatial working memory deficits, particularly among the female progeny ( He et al., 2006a ). Consistent with these results, the offspring of cocaine-exposed rat sires showed increased perseverance in a  T-maze learning task ( Abel et al., 1989 ). Interestingly, the expression of DNA  methyltransferase  1 (Dnmt-1) was decreased in the  seminiferous tubules  of the testis (the locus of spermatogenesis) of sires that self-administered cocaine ( He et al., 2006a ). Given that DNA  methyltransferases  play a critical role in maintaining imprinting in  germ cells , reduced Dnmt-1 in the sperm of sires that self-administered cocaine is a potential mechanism that may account for the intergenerational influence of paternal cocaine exposure ( He et al., 2006a ).

Notably, none of these studies examined cocaine reinforcement among the offspring. Therefore, two of the authors of the current paper (Vassoler and Pierce) and our colleagues examined cocaine self-administration in the offspring of sires that self-administered cocaine for 60 days. The male offspring of cocaine-experienced sires acquired cocaine self-administration more slowly and had decreased levels of cocaine intake relative to controls. Cocaine self-administration in female offspring did not differ between cocaine- and saline-exposed sires ( Vassoler et al., 2013 ). Previous work indicated that increased  brain derived neurotrophic factor  ( BDNF ) in the  medial prefrontal cortex  ( mPFC ) blunted the behavioral effects of cocaine ( Berglind et al., 2007  and  Sadri-Vakili et al., 2010 ). We showed that mPFC Bdnf mRNA and protein were increased only in the male offspring of sires that self-administered cocaine (  Vassoler et al., 2013 ). Moreover, increased association of  acetylated   histone H3  with Bdnf promoters was observed in the mPFC of male offspring, which is one mechanism that may underlie the enhanced BDNF transcription in the mPFC of cocaine-sired rats. Systemic administration of a BDNF  receptor antagonist  (the  TrkB receptor  antagonist ANA-12) normalized the decreased cocaine self-administration in male cocaine-sired rats. In addition, the association of acetylated histone H3 with Bdnf promoters was increased in the sperm of sires that self-administered cocaine (  Vassoler et al., 2013 ). Taken together, these findings indicated that voluntary paternal ingestion of cocaine reprograms the germline resulting in enhanced BDNF expression in the mPFC among male progeny, which appears to confer resistance to the reinforcing effects of cocaine. This result should be interpreted cautiously. It is as yet unknown if these male offspring have impairments in reward and motivational circuits that underlie the decreased self-administration behavior.

We examined paternal transmission in order to avoid the influence of in utero cocaine exposure and the potential influence of prior cocaine experience by dams on maternal behavior. It is possible that even the relatively brief exposure to a cocaine-experienced male during breeding might have an impact on maternal behavior. We examined licking/grooming and other maternal behaviors and found no differences between the dams bred with cocaine-experienced sires relative to controls (  Vassoler et al., 2013 ).

A critical unanswered question is whether this cocaine resistance  phenotype  is present in the F2 generation (i.e. the grandoffspring of cocaine-experienced sires). Since the sperm of the sires was exposed to cocaine, it cannot be assumed that the heritability is transgenerational. Indeed, as suggested above, it is possible that cocaine binding to the spermatozoa could allow for transport of cocaine into the fertilized ovum ( Yazigi et al., 1991 ). In order to demonstrate transgenerational heritability in these experiments, the epigenetic markers and associated  phenotypes  need to be observed in the F2 generation. It also is unclear why only the male offspring found cocaine less reinforcing. Given that  ovarian hormones  have been shown to have profound influences on the behavioral effects of cocaine ( Anker and Carroll, 2011 ,  Hu and Becker, 2008  and  Quinones-Jenab and Jenab, 2010 ), it is possible that direct or indirect influences on  steroid hormones  may underlie the observed gender differences in the offspring of cocaine-experienced sires.

Resistance to cocaine in the offspring of sires that self-administered this drug is apparently at odds with human epidemiological data indicating that cocaine addiction is highly heritable ( Kendler et al., 2007 ,  Merikangas et al., 1998  and  Tsuang et al., 1998 ). Of course, cocaine addiction is not completely heritable and all addictions are influenced by drug availability and many other environmental factors. In our study, environmental influences were controlled to an extent that is impossible in epidemiological experiments.

7. Conclusions

The idea that the environment of one generation can impact the  phenotype  of subsequent generations is not new. Jean-Baptiste Lamarck proposed a theory of evolution that incorporated environmental conditions as well as reproductive fitness into hereditary changes. His work was largely discredited and ignored for lack of mechanism and in favor Darwin's elegant description of “survival of the fittest” ( Burkhardt, 2009 ). However, it seems that some of his ideas had merit and can now be explained by the phenomenon of  epigenetics . The way that environmental exposures of one generation can influence and affect subsequent generations is of critical importance to the understanding of human behavior and evolution.

This review demonstrated that there are many consequences for the offspring of parents that were exposed to drugs, even in the absence of direct fetal exposure (summarized in Table 1 ). The idea that environmental toxins ingested prior to conception by either parent can have such a pronounced impact on the offspring needs further research and should garner more attention. In fact, one study found that the effects of paternal and maternal drug use had an additive effect on offspring's adolescent drug use trajectory rather than an interactive parental effect ( Walden et al., 2007 ). This suggests the need for policy change and new campaigns to spread awareness among men and women during adolescence and throughout child bearing years.

Table 1.

Preconception effects of drug exposure on subsequent generation.

Table options

The results presented here suggest the intriguing, and potentially alarming, possibility that exposure to drugs of abuse produce transmissible  epigenetic  changes that result in profound alterations to the physiology and behavior of offspring. However, for an effect to be truly non-genomic  epigenetic inheritance , many of the experiments need to be carried to further generations to avoid exposure of the  germ cells  to the drug of abuse. The few studies that have looked beyond the first generation suggest that many  phenotypes persist. Regardless of the number of future generations preconception drug use influences, the impact on first generation offspring alone is sufficient to justify further research defining the extent of epigenetic heritability of phenotypes associated with parental drug abuse and the specific mechanisms underlying these effects

europharmacology

Volume 76, Part B, January 2014, Pages 269–275

NIDA 40th Anniversary Issue

125461||

Invited review

The impact of exposure to addictive drugs on future generations: Physiological and behavioral effects

· F.M. Vassoler a, , , 

· E.M. Byrnes a, 

· R.C. Pierce b

 Show more

doi:10.1016/j.neuropharm.2013.06.016

Get rights and content

Highlights

•

Transgenerational consequences of exposure to drugs of abuse.

•

Paternal and maternal transmission to offspring prior to conception.

•

Alcohol, nicotine, cannabinoids, opioids, cocaine.

Abstract

It is clear that both genetic and environmental factors contribute to drug addiction. Recent evidence indicating trans-generational influences of drug abuse highlight potential epigenetic factors as well. Specifically, mounting evidence suggests that parental ingestion of abused drugs influence the physiology and behavior of future generations even in the absence of prenatal exposure. The goal of this review is to describe the trans-generational consequences of preconception exposure to drugs of abuse for five major classes of drugs: alcohol, nicotine, marijuana, opioids, and cocaine. The potential epigenetic mechanisms underlying the transmission of these phenotypesacross generations also are detailed.

This article is part of a Special Issue entitled ‘NIDA 40th Anniversary Issue’.

Keywords

· Transgenerational; 

· Epigenetic; 

· Drugs of abuse; 

· Alcohol; 

· Nicotine; 

· Marijuana; 

· Opioids;

· Cocaine

1. Introduction

Drug addiction is a serious medical and social issue in the United States and around the world. There is consistent evidence that substance use disorders run in families (Bierut et al., 1998, Brook et al., 2002, Cloninger et al., 1981 and Merikangas et al., 1998). Adoption, twin, and sibling studies implicate genetic factors in the heritability of abuse (Cloninger et al., 1981). However, simple genetic mechanisms of inheritance cannot explain all results (Cloninger et al., 1981 and Schuckit et al., 1972). Societal differences in drug use and consumption patterns vary from different time periods and between countries suggesting a large environmental component (UNODC, 2012). Thus, vulnerability to develop an addiction is dependent on both genetics and the environment. The sum of separate genetic and environmental contributions cannot fully explain the heritability either. Therefore, the interaction between genetics and the environment may help explain some of the discrepancies (Cloninger et al., 1981).

Epigenetics is a key mechanism by which the environment can influence and interact with genetics. In recent years, the term epigenetics has been used to describe myriad processes (Haig, 2004). For example, modifications to the structure of chromatin or DNA without changes in the sequence that affect gene transcription even in non-dividing cells such as DNA methylation or histone acetylation are described as epigeneticmodifications (Holliday, 1989). However, some definitions of epigenetics emphasize that the modifications in gene expression that do not involve alterations in the DNA sequencemust be heritable, spanning multiple generations. There has recently been an increase in studies examining the transgenerational effects of environmental toxins on offspring. The goal of the current review is to examine the available literature regarding the effects on offspring of parental drug exposure in the absence of any direct fetal exposure for five major drugs of abuse. While offspring susceptibility to drug use is of particular interest, all behavioral, molecular, and physiological changes in the offspring will be reported. The majority of the studies available on this topic focus on paternal transmission of epigeneticphenotypes, as this model eliminates any direct fetal exposure and ostensibly avoids maternal rearing effects. With regard to maternal transmission, there is certainly an extensive body of literature documenting offspring effects following prenatal exposure to drugs of abuse (Malanga and Kosofsky, 2003 and Sithisarn et al., 2012). Due to the possible direct effects of in utero exposure on the fetus, as well as the numerous confounds that drug use during pregnancy introduces (e.g. changes in nutritional status, renal function, vascular perfusion, etc.), prenatal substance use models will not be included in the current review. We will, however, include data from studies examining transgenerational effects of female drug use occurring prior to conception. Thus, the current review will examine the effects of parental exposure to drugs of abuse prior to conception on the development of subsequent generations.

2. Alcohol

Currently, alcohol is the most commonly abused drug in the United States. In 2011, the center for disease control estimated that 60% of males and 44% of females engage in chronic alcohol drinking (Edward J. Sondik et al., 2012). While the neurobehavioral effects of fetal alcohol exposure are well described, less is known about the effects of parental exposure to alcohol prior to conception. It should be noted, however, that reports and writings as early as the 1720's, during the so-called gin epidemic, observed that both maternal and paternal alcohol use had detrimental effects on offspring (Warner and Rosett, 1975). While few studies have examined maternal alcohol use prior to pregnancy (i.e. in the absence of prenatal use), several findings have demonstrated effects of paternal alcohol exposure on offspring development. Indeed, as early as 1913 animal studies found that offspring sired by alcohol inhaling rats demonstrated malformations, low birth weight, retarded growth and increased neonatal mortality across several generations (Friedler, 1996). Clearly the concept that alcohol use by the father, and not solely his genetic composition, can affect future progeny is not novel. The interest in paternal effects, however, has been reinvigorated by the emergence of the field of epigenetics, with a number of preclinical findings suggesting transgenerationalepigenetic effects of paternal alcohol use.

Several of the initial animal studies on paternal alcohol effects focused on basic parameters of reproductive success, such as fertility and fecundity, following exposure to alcohol in peripubertal males. These studies observed a significant reduction in the number of successful pregnancies which decreased from 92% in controls to 75% in naïve females mated with alcohol-drinking sires (Emanuele et al., 2001). Litter size was also substantially reduced (Cicero et al., 1990 and Emanuele et al., 2001). It was determined that alcohol exposure during puberty modified sexual maturation and resulted in decreased testes and secondary sex organ weight, eliminated the typical pubertal surge in testosterone, decreased beta-endorphin levels in the hypothalamus, and enhanced testicular oxidative injury (Cicero et al., 1990 and Emanuele et al., 2001). Interestingly, offspring of these pubertal alcohol-exposed sires demonstrated similar alterations, including decreased serum testosterone levels, reduced seminal vesicleweights, and lower levels of hypothalamic beta-endorphin (Abel and Lee, 1988 and Cicero et al., 1990).

Alcohol-sired offspring also demonstrate abnormalities in development. In an elegantly designed rodent study, Jamerson and colleagues revealed that paternal alcohol exposure that was ongoing, or that had ceased weeks prior to conception resulted in more rapid development of various reflexes, differences in gait, and thicker cortical layers(Jamerson et al., 2004). It was also noted that timing of alcohol exposure in relation to conception impacted the neurobehavioral effects of the offspring (Jamerson et al., 2004). For example, one study found that a single exposure to alcohol just prior to conception resulted in a significant increase in small for gestational age offspring as well as an increase in offspring demonstrating significant malformations (Bielawski and Abel, 1997). Moreover, alcohol-exposed sires also produced offspring displaying increasedadrenal cortex and decreased spleen weights (Abel, 1993b). Finally, there is evidence that metabolic and immune functioning may be disrupted in alcohol-sired offspring, given that they display reductions in leptin levels (Emanuele et al., 2001) and a diminished immune response (Berk et al., 1989 and Hazlett et al., 1989).

In terms of behavioral effects, both increases and decreases in activity have been noted in offspring of alcohol-consuming sires (Abel, 1989a, Abel, 1989b, Abel, 1993a, Abel, 1993b and Abel and Lee, 1988), with the direction of these effects mediated by a number of factors including the level of alcohol consumption, the time between exposure and conception (Jamerson et al., 2004), and the age at the time of testing (Abel, 1989a). Alterations in behavioral activity following amphetamine were also noted in male offspring (Abel, 1993a), with some evidence suggesting that increased activity was dependent on the cholinergic system (Abel, 1994). Potential modifications in the cholinergic system of alcohol-sired offspring are notable given that deficits in learning and memory have also been reported. For example, offspring sired by alcohol treated males demonstrated impairments in spatial learning (Wozniak et al., 1991) and had increased latencies to reach a choice point in a T-maze (Abel, 1994 and Abel and Lee, 1988). In addition to deficits observed in males, alcohol-sired female offspring showed impaired performance in a two-way shock avoidance learning task (Abel and Tan, 1988). Finally, offspring of alcohol-consuming sires demonstrated decreased grooming as well as decreased immobility in a forced swim test, an effect that was rescued by imipramineand propranolol and exacerbated by yohimbine and metergoline (Abel, 1991a, Abel, 1991b and Abel and Bilitzke, 1990). Together, these results indicate a detrimental behavioral phenotype of paternal alcohol consumption on both male and female offspring.

Examining epigenetic parental effects in human populations can be difficult. Human studies have, however, revealed correlations between parental drinking behavior and offspring initiation and drinking patterns. Thus, heavy paternal drinking or heavy-episodic drinking in both parents predicts earlier onset of offspring drinking as well as heavier drinking (Vermeulen-Smit et al., 2012). Children born with characteristics associated with fetal alcohol syndrome whose mother's did not drink but fathers were alcoholics have also been observed (Lemoine et al., 2003). Additionally, there is evidence that sons of early-onset (adolescent) alcoholic fathers perform more poorly on tests of verbalintelligence and attention than late onset (adult) alcoholics (Tarter et al., 1989). While this may suggest that adolescence, or periods of significant neural and endocrinedevelopment, are particularly important in determining the effects of paternal alcohol exposure, one cannot exclude the possibility that early-onset drinking may be a marker of genetic vulnerability or additional comorbid pathologies. Finally, in addition to deficits in cognition, attention, and visuospatial capacity observed in children of alcoholic fathers, increased hyperactivity has also been noted (Goodwin et al., 1975), although the relationship between hyperactivity and paternal alcohol consumption remains equivocal (Knopik et al., 2009). Much less work has been done examining the effects of maternal preconception alcohol consumption. However, a decreased birth weight has been observed in children of alcoholic women that abstain from use during pregnancy (Little et al., 1980, Livy et al., 2004 and Ramsay, 2010). While human studies remain more difficult to interpret than preclinical research due to many aspects that are impossible to control, it is important that both lines of research continue in order to fully understand the scope of preconception alcohol use and abuse.

3. Nicotine

Nicotine is the second most abused substance in the United States. In 2011 it was reported that 21.6% of adult men and 16.5% of adult women smoke regularly (Agaku et al., 2011). It is clear from both human epidemiological and preclinical research models that exposure to tobacco smoke in utero is harmful to development and can result in low birth weight, sudden infant death syndrome, and serious behavioral issues in the offspring ( Abbott and Winzer-Serhan, 2012). Examining the indirect effects (i.e. non-prenatal exposure) in humans is more challenging because it is difficult to distinguish the indirect epigenetic effects of paternal or maternal nicotine abuse on the offspring from potential exposure to second-hand smoke in utero or post-partum. Keeping that in mind, some studies have shown increased risk of spontaneous abortion associated with paternal smoking ( Blanco-Munoz et al., 2009 and Venners et al., 2004). However, others have shown no association ( Chatenoud et al., 1998, Windham et al., 1992 and Windham et al., 1999). Moreover, paternal smoking is a factor associated with anorectal malformations in humans in multiple studies (Zwink et al., 2011). Human epidemiological studies also indicate that early paternal smoking (onset before age 11) is associated with greater body mass index in male, but not female, offspring ( Pembrey, 2010 and Pembrey et al., 2006).

Aside from these few studies looking at the effect of paternal smoking on offspring, numerous studies have found evidence that male smoking affects multiple fertility factors in the male. Thus, paternal tobacco use is associated with defects in the tail of the spermatozoon (Ozgur et al., 2005) and decreases in sperm density, motility, and morphology (Sofikitis et al., 1995). Another study found no differences in sperm concentration or motility but decreased semen volume (Pasqualotto et al., 2006). Taken together, these studies point toward decreases in male fertility and/or embryo viability which suggests potential negative effects in the offspring. To date, we found only one animal study which found that paternal nicotine smoke exposure did not affect development or any behavioral parameters measured (Gaworski et al., 2004). More animal work is needed to determine if nicotine exposure and smoking has transgenerational epigenetic effects.

4. Cannabinoids

Marijuana is the most widely abused illicit substance in the United States. Of particular concern is the high percentage of adolescents that engage in marijuana use. It is estimated that almost 50% of teenagers have used marijuana and approximately 9% use marijuana heavily (>20 days out of the month) (Metlife, 2011). Marijuana use by adolescents may be particularly problematic as developing systems may be more vulnerable to the impact of exogenous cannabinoids (Rice and Barone, 2000). Abuse by adolescent females is of particular concern as the endogenous cannabinoid system is important for reproductive physiology, with CB1 receptors as well as endogenous cannabinoids expressed in the ovaries, uterine endometrium, and other peripheralendocrine tissue (Bari et al., 2011). Thus, cannabinoid exposure during this critical period could result in lasting modifications in female reproduction, and might impact future offspring. To examine the potential effects on offspring, adolescent females were exposed to cannabinoids during adolescence. They were then maintained drug-free for over 3 weeks and bred during adulthood. Using this model, it was shown that both adolescent and adult male offspring of adolescently-exposed dams exhibit greater sensitivity to morphine-induced conditioned place preference than the control animals, even in the absence of any direct in utero exposure ( Byrnes et al., 2012). Moreover, unpublished data from the Byrnes laboratory shows that female offspring demonstrate enhanced expression of morphine-induced locomotor sensitization accompanied by increased expression of the mu opioid receptor in the nucleus accumbens.

While no studies to date have examined the effect on the offspring of cannabinoid exposure in the sire prior to conception, available evidence suggests that disruptions in endocrine functioning resulting from cannabinoid exposure may cause deleterious effects in subsequent generations. For example, tetrahydrocannabinol (THC) disrupts gonadal functions by depriving testicular cells of energy reserves and stimulatingandrogen-binding protein secretion which leads to oligospermia in chronic cannabissmokers. THC also interferes with production of prolactin, luteinizing hormone (LH), andfollicle stimulating hormone (FSH), which causes reduced testosterone production in the testes (Banerjee et al., 2011, Harclerode, 1984 and Husain and Khan, 1985). It has also been suggested that acute cannabinoid treatment affects the quality and quantity of spermatozoa produced by the testis (Harclerode, 1984). Animal models have demonstrated that cannabinoid administration suppresses gonadal steroids, growth hormone, prolactin, thyroid hormone and activates the HPA-axis (Banerjee et al., 2011 and Brown and Dobs, 2002). These effects have been shown to be mediated by the binding of exogenous cannabinoids to the endogenous receptor in or near thehypothalamus (Brown and Dobs, 2002). However, the effects in humans have been inconsistent likely due to the development of tolerance (Brown and Dobs, 2002). Finally, in humans, two case controlled studies found an increased risk for congenital heart defects associated with paternal marijuana use (Ewing et al., 1997 and Wilson et al., 1998). Clearly, there is a large gap in the literature examining the effects of marijuana use preconception on future generations.

5. Morphine

The significant impact of maternal opioid use on fetal outcomes and offspring neurodevelopment has been an area of interest for decades (Hutchings, 1982, Malanga and Kosofsky, 1999 and Vathy, 2002). In both substance abusing women and animal models of prenatal substance use the fetus is directly exposed to opioids, limiting the applicability to a discussion on transgenerational epigenetic processes. To avoid direct fetal exposure, animal models were developed that expose both males and females to opioids preconception and then examine the effects on their offspring. Such studies have primarily examined the impact of exposure to morphine during adolescence. In male rats, morphine administered throughout adolescent development resulted in significant modifications in sexual maturation in the males themselves. When these males were mated to drug naïve females several weeks after morphine exposure, they produced smaller litters and as adults their offspring demonstrated significant modifications inendocrine parameters, including sex specific changes in adrenal weights, luteinizing hormone and hypothalamic β-endorphin (Cicero et al., 1991). As all that the male contributed in this model was sperm, such findings strongly suggest that prior exposure to opioids can induce transgenerational epigenetic effects even in the absence of continuing use.

Similar findings have been revealed using a female rat model. In those studies, females were exposed to an intermittent, increasing dose regimen of morphine during early-mid adolescence. All animals were then drug-free for several weeks prior to mating with drug naïve males. In these studies there were no significant differences in any postnatal parameter (i.e. litter size, weight, sex ratio). When offspring were examined as adults, however, a number of significant effects have been observed. These include sex-specific changes in social and emotional behaviors (Byrnes et al., 2011 and Johnson et al., 2011) as well as alterations in their response to morphine (Byrnes, 2005 and Byrnes et al., 2011). More recent findings using this model also observed significant changes in the functional response to dopamine agonists coupled with increased levels of dopamine- and opioid-related gene expression in the nucleus accumbens. Of note, these effects were observed in both the first (F1) and second (F2) generation, suggesting multigenerational epigenetic effects triggered by adolescent exposure (Byrnes et al., 2013).

The mechanisms underlying such intergenerational transmission are unknown and perhaps more complicated when considering maternal transmission. Exposure to drugs of abuse, even when experienced prior to conception, may alter the prenatal hormonal milieu, thereby modifying the developmental trajectory of offspring. Alternatively, or perhaps additionally, modifications in maternal care as a result of prior exposure to opioids, could impact the development of offspring. Indeed, morphine administered both prior to parturition or during active mothering significantly alters maternal care (Bridges and Grimm, 1982, Mann et al., 1991, Miranda-Paiva et al., 2001 and Slamberova et al., 2001). Moreover, non-genomic transmission via subtle variations in maternal care have been demonstrated repeatedly in animal models (McLeod et al., 2007, Szyf et al., 2007,Weaver, 2007 and Weaver et al., 2004). Thus, one possible mechanism of transmission may be alterations in the pre- and/or postnatal environment.

Alternatively, exposure to opioids could directly affect epigenetic germline cells in the exposed male and female. Such effects could alter early embryogenesis and impact a number of developmental processes. These effects could then be passed forward via alterations in behavior/physiology of the next generation or via direct epigenetic inheritance. Additional studies are needed to determine the mechanisms underlying transmission in both males and females exposed to opioids. Nonetheless, the nature of the observed offspring effects, including dysregulation of the hypothalamic–pituitary–adrenal axis, increased sensitivity to opioids, and blunted response to dopamine agonists, all suggest that parental exposure to opioids prior to conception may increase the risk of substance use in their future offspring.

6. Cocaine

In terms of parental cocaine administration, the clinical and preclinical literatures have focused primarily on the developmental effects of prenatal cocaine exposure. In that regard, preclinical studies relatively consistently show disrupted cortical andhippocampal development in the offspring as well as impaired executive function and memory following in utero cocaine exposure ( Dow-Edwards, 2011, Lidow, 2003 and Malanga and Kosofsky, 2003). Epigenetic mechanisms appear to play a role in these changes in that male mice exposed to cocaine prenatally had altered patterns ofDNA methylation in hippocampal pyramidal neurons (Novikova et al., 2008), which could underlie impaired sustained attention and spatial working memory observed in these offspring (He et al., 2006b). While interesting, studies of this sort are directly examining the effect of cocaine on development. Moreover, the dams were exposed to cocaine, which may have influenced maternal behaviors. These issues are largely obviated in experiments in which paternal cocaine exposure is examined.

It has long been known that cocaine concentrates in the testes at levels second only to the brain (Mule et al., 1977 and Yazigi et al., 1991) due primarily to specific binding sites in spermatozoa (Yazigi et al., 1991). Animal studies showed that prolonged experimenter-administered cocaine (72–150 days) impaired spermatogenesis (George et al., 1996) and increased the percentage of sperm with tails separated from their heads (Abel et al., 1989). Although some evidence indicated that the fertility of cocaine-treated rats was substantially impaired (George et al., 1996), fecundity remained sufficient to examine the effects of paternal cocaine exposure on their offspring. For example, paternal cocaine exposure decreased the weight of their progeny (George et al., 1996 and Killinger et al., 2012) but see also (Abel et al., 1989). Experimenter-delivered cocaine to mouse sires also resulted in increased immobility in the tail suspension test, a model of depression, but had no effect on locomotor activity, measures of anxiety or learning and memory (Killinger et al., 2012). In contrast, the offspring of mouse sires that self-administered cocaine displayed attention and spatial working memory deficits, particularly among the female progeny (He et al., 2006a). Consistent with these results, the offspring of cocaine-exposed rat sires showed increased perseverance in a T-mazelearning task (Abel et al., 1989). Interestingly, the expression of DNA methyltransferase 1 (Dnmt-1) was decreased in the seminiferous tubules of the testis (the locus of spermatogenesis) of sires that self-administered cocaine (He et al., 2006a). Given that DNA methyltransferases play a critical role in maintaining imprinting in germ cells, reduced Dnmt-1 in the sperm of sires that self-administered cocaine is a potential mechanism that may account for the intergenerational influence of paternal cocaine exposure (He et al., 2006a).

Notably, none of these studies examined cocaine reinforcement among the offspring. Therefore, two of the authors of the current paper (Vassoler and Pierce) and our colleagues examined cocaine self-administration in the offspring of sires that self-administered cocaine for 60 days. The male offspring of cocaine-experienced sires acquired cocaine self-administration more slowly and had decreased levels of cocaine intake relative to controls. Cocaine self-administration in female offspring did not differ between cocaine- and saline-exposed sires (Vassoler et al., 2013). Previous work indicated that increased brain derived neurotrophic factor (BDNF) in the medial prefrontal cortex (mPFC) blunted the behavioral effects of cocaine (Berglind et al., 2007 and Sadri-Vakili et al., 2010). We showed that mPFC Bdnf mRNA and protein were increased only in the male offspring of sires that self-administered cocaine ( Vassoler et al., 2013). Moreover, increased association of acetylated histone H3 with Bdnf promoters was observed in the mPFC of male offspring, which is one mechanism that may underlie the enhanced BDNF transcription in the mPFC of cocaine-sired rats. Systemic administration of a BDNF receptor antagonist (the TrkB receptor antagonist ANA-12) normalized the decreased cocaine self-administration in male cocaine-sired rats. In addition, the association of acetylated histone H3 with Bdnf promoters was increased in the sperm of sires that self-administered cocaine ( Vassoler et al., 2013). Taken together, these findings indicated that voluntary paternal ingestion of cocaine reprograms the germline resulting in enhanced BDNF expression in the mPFC among male progeny, which appears to confer resistance to the reinforcing effects of cocaine. This result should be interpreted cautiously. It is as yet unknown if these male offspring have impairments in reward and motivational circuits that underlie the decreased self-administration behavior.

We examined paternal transmission in order to avoid the influence of in utero cocaine exposure and the potential influence of prior cocaine experience by dams on maternal behavior. It is possible that even the relatively brief exposure to a cocaine-experienced male during breeding might have an impact on maternal behavior. We examined licking/grooming and other maternal behaviors and found no differences between the dams bred with cocaine-experienced sires relative to controls ( Vassoler et al., 2013).

A critical unanswered question is whether this cocaine resistance phenotype is present in the F2 generation (i.e. the grandoffspring of cocaine-experienced sires). Since the sperm of the sires was exposed to cocaine, it cannot be assumed that the heritability is transgenerational. Indeed, as suggested above, it is possible that cocaine binding to the spermatozoa could allow for transport of cocaine into the fertilized ovum (Yazigi et al., 1991). In order to demonstrate transgenerational heritability in these experiments, the epigenetic markers and associated phenotypes need to be observed in the F2 generation. It also is unclear why only the male offspring found cocaine less reinforcing. Given that ovarian hormones have been shown to have profound influences on the behavioral effects of cocaine (Anker and Carroll, 2011, Hu and Becker, 2008 and Quinones-Jenab and Jenab, 2010), it is possible that direct or indirect influences on steroid hormones may underlie the observed gender differences in the offspring of cocaine-experienced sires.

Resistance to cocaine in the offspring of sires that self-administered this drug is apparently at odds with human epidemiological data indicating that cocaine addiction is highly heritable (Kendler et al., 2007, Merikangas et al., 1998 and Tsuang et al., 1998). Of course, cocaine addiction is not completely heritable and all addictions are influenced by drug availability and many other environmental factors. In our study, environmental influences were controlled to an extent that is impossible in epidemiological experiments.

7. Conclusions

The idea that the environment of one generation can impact the phenotype of subsequent generations is not new. Jean-Baptiste Lamarck proposed a theory of evolution that incorporated environmental conditions as well as reproductive fitness into hereditary changes. His work was largely discredited and ignored for lack of mechanism and in favor Darwin's elegant description of “survival of the fittest” (Burkhardt, 2009). However, it seems that some of his ideas had merit and can now be explained by the phenomenon of epigenetics. The way that environmental exposures of one generation can influence and affect subsequent generations is of critical importance to the understanding of human behavior and evolution.

This review demonstrated that there are many consequences for the offspring of parents that were exposed to drugs, even in the absence of direct fetal exposure (summarized inTable 1). The idea that environmental toxins ingested prior to conception by either parent can have such a pronounced impact on the offspring needs further research and should garner more attention. In fact, one study found that the effects of paternal and maternal drug use had an additive effect on offspring's adolescent drug use trajectory rather than an interactive parental effect (Walden et al., 2007). This suggests the need for policy change and new campaigns to spread awareness among men and women during adolescence and throughout child bearing years.

Table 1.

Preconception effects of drug exposure on subsequent generation.

Table options

The results presented here suggest the intriguing, and potentially alarming, possibility that exposure to drugs of abuse produce transmissible epigenetic changes that result in profound alterations to the physiology and behavior of offspring. However, for an effect to be truly non-genomic epigenetic inheritance, many of the experiments need to be carried to further generations to avoid exposure of the germ cells to the drug of abuse. The few studies that have looked beyond the first generation suggest that many phenotypespersist. Regardless of the number of future generations preconception drug use influences, the impact on first generation offspring alone is sufficient to justify further research defining the extent of epigenetic heritability of phenotypes associated with parental drug abuse and the specific mechanisms underlying these effects