Online Issues

Journal Vol 47(1)

Introduction

Knowledge Gained from Animal Studies of the Fetus and Newborn: Application to the Human Premature Infant

Estelle B. Gauda

Estelle B. Gauda, M.D., is an Associate Professor of Pediatrics, Division of Neonatology, Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, Maryland.

Key Words: animal models of the fetus and newborn; chronic lung disease; control of breathing; human cytomegalovirus; limit of viability; neonatal brain injury; neonatal opiate dependence; premature infant

Advances in care of the fetus and newborn have resulted in the survival of infants born at slightly more than half of gestation. Rapid technological advances have led to therapies resulting in a marked increase in the survival of the youngest of infants. The survival of these infants has also increased the awareness of individuals in the medical and research communities that there is a limit to viability. Despite the most advanced technologies in respiratory, nutritional, and cardiovascular care, this population of infants is still at great risk for mortality and morbidity. The interplay between the development of immature organs, exposure to a suboptimal intrauterine environment, or premature exposure to an extrauterine environment, and resulting respiratory and neurological outcomes have generated a plethora of research to understand this interplay better and to optimize the outcomes. It is extremely difficult to perform studies in human infants because of the multitude of ethical issues and risk associated with unproven therapies in this vulnerable population. Thus, well-conducted studies in animal models are particularly relevant and essential to improving the care of the fetus and newborn. It should be noted that differences between altricial and precocial species are particularly important in animal models of the fetus and newborn. The periods of vulnerability to the same organ systems might be prenatal in sheep, nonhuman primates and humans (atricial animals) while it will be postnatal in rodents (precocial). The rodent, thus, can be an easier model to study normal development, plasticity, and pathogenesis of disease in the premature infant.

For this issue of ILAR Journal, I have invited experts in their corresponding fields to discuss how responsible use of animals in research has led to the unraveling of the interplay between the developing fetus and newborn and their environments. The most common morbidities that reduce the quality of life in infants affect the brain and lung, and result in major neurological handicaps in motor and sensory function, cognition, and chronic lung disease. Elegant studies in fetal and newborn animals have increased our understanding of the mechanisms of brain injury in term and preterm infants, of respiratory distress syndrome, and of resulting chronic lung disease. Recent work in fetal and newborn models has begun to give us a better understanding of some of the biological mechanisms that explain the epidemiological association between intrauterine growth retardation and resulting morbidities in adults such as hypertension, cardiovascular diseases, obesity, and diabetes. This knowledge has not only increased our understanding but has also led to therapies for prevention and treatment to optimize outcomes for the smallest of infants.

The article by Dr. Festing (2006) leads this issue, with pertinent advice that cannot be overemphasized: ". . . experiments involving neonates should be unbiased, be powerful, have a good range of applicability, not be excessively complex, and be statistically analyzable to show the range of uncertainty in the conclusions." However, the author reminds us that to achieve these goals, it is necessary to apply certain rules to multiparous animals such as rabbits, mice, and rats. An investigator must identify the experimental unit and take litter effects into account for the results of the experiment to yield interpretable conclusions. Dr. Festing walks the reader through the process of determining the experimental unit when designing experiments using multiparous animals, and through statistically analyzing the data. Although we frequently think of randomization as it applies to studies in humans, randomization is also an important study design method that should be performed routinely in studies using animals.

Large animals (e.g., lambs, piglets, neonatal cats, and neonatal dogs) have been used routinely in physiological studies to characterize maturation of central and peripheral neuronal networks that control respiration. Because of the ease of obtaining physiological measurements, these animal models have been particularly useful in characterizing maturation of respiratory and cardiovascular control during fetal life. Fetal breathing, which is intermittently characterized by frequent apnea, is regulated indirectly by oxygen tension (hypoxia increases frequency of apnea). However, with maturation, fetal breathing becomes stable, and at term, apnea is infrequent. Nevertheless, despite sustained breathing with maturation, infants born prematurely have frequent apneas associated with bradycardic and hemoglobin desaturation events. Infants born prematurely still have a high risk for sudden infant death, and some infants born at term have abnormalities in breathing that necessitate continued ventilatory management (central hypoventilation syndrome).

Although technically more challenging, rodents have been increasingly used in the study of maturation of respiratory control. Ventilatory responses to hypoxia and hypercapnia in rats and mice have a similar developmental pattern to the human premature infant and other newborn mammalian animal models that have been used in physiological studies. Among the specific characteristics of the maturation of breathing in newborns are the well-described resetting of peripheral arterial chemoreceptors during the first few days after birth, the characteristic biphasic ventilatory response to hypoxia, and the incremental increase to in ventilation in response to hypercapnia with maturation. Dr. Gaultier and colleagues (2006) have been on the forefront in developing a noninvasive technique (whole-body flow barometric plethysmography) to study the maturation of breathing in newborn mice. One important challenge in using small newborn rodents is that metabolism is closely coupled to ventilatory responses, thus temperature and metabolism must be measured in these animals who weigh as little as 5 g at birth. Dr. Gaultier and coauthors report being able to detect arousals in newborn mice as early as the first postnatal day. This newly characterized mouse model is robust in that it allows for assessment of breathing patterns in genetically altered mice. These authors provide a valuable overview of specific breathing patterns in mutant mice. This mouse model has allowed scientists to begin to interrogate which genes may be involved in the development and regulation of respiratory control. By combining information provided by studies in mutant mice and linkage studies performed in humans, several disorders of respiratory control (e.g., congenital central hypoventilation syndrome and Prader-Willi syndrome) have been linked to specific genes.

Multiparous animals have been used extensively to understand genes that regulate the development of the lung. Much information has been learned from transgenic mouse models, and it has led to an improved understanding and characterization of lung disease in the premature and mature infant. Drs. Bridges and Weaver (2006) describe selected transgenic mouse models that are used to elucidate the molecular pathways that regulate key prenatal, perinatal, and postnatal morphogenetic events involved in lung development and cellular differentiation during the prenatal period; in lung function during the perinatal period; or in lung homeostasis during the postnatal period. The discovery of surfactant deficiency as a major cause of mortality in infants born prematurely, along with the resultant plethora of animal and human studies, has led to a dramatic improvement in the care and survival of the youngest of infants. In addition to the ground-breaking research in newborn animals that led to identifying the pathogenesis (e.g., surfactant deficiency) that is the leading cause of respiratory failure in premature infants, newborn animals continue to be the source of surfactant (bovine and porcine) for the treatment of surfactant deficiency in premature infants. Drs. Bridges and Weaver carefully interweave basic histology and morphology of lung and vascular development with evidence from transgenic mouse models that supports specific genes involved in the process. Although some of the genetically manipulated mouse models are embryonically lethal, conditional knockdown experimental methods have led to a more robust experimental design to interrogate the role of certain genes in the development of chronic lung disease and the interaction between the extrauterine environment (e.g., oxygen tension, postnatal steroids, infectious agents), which predispose the developing lung to injury and altered repair.

Although much information can be obtained from a number of animal models, the article by Dr. Northington (2006) emphasizes the importance of determining whether the model is similar to the human for the specific system being modeled. Brain development does not follow the same time trajectory across species, and the maturational profile may differ between brain regions across species. Thus, it is important to understand cross-species histological and functional maturational profile of specific brain regions when selecting an animal model of neonatal brain injury. The mostly widely used model of brain injury (Rice-Vannuci) is the newborn rat at 7 postnatal days of age because this model has the brain maturity equivalent to that of an early third trimester human fetus. Thus it is a model of brain injury that occurs in the premature human. The Rice-Vannuci model has also been adapted for the newborn mouse, which differs by 1 day in brain maturation compared with the rat. In contrast to these rodent models, early third trimester human brain equivalence occurs at approximately 90 to 110 days gestation in the fetal sheep, which has a gestation of 140 days.

The animal models described above have been very instrumental in understanding mechanisms (including specific genes and signaling pathways) that result in necrosis and apoptosis in the immature brain. However, these models lack similarity to the functional sequelae seen in the human premature and term infant as a result of hypoxic ischemia brain injury. A new neonatal rabbit model of brain injury in the immature rabbit appears to hold considerable promise because the animals with induced brain injury also display abnormalities in motor control similar to what has been described for white matter injuries in the premature infant. These models of neonatal brain injury also will allow a better assessment of potential therapeutic modalities that could ultimately benefit infants, as described for hypothermic treatment in Dr. Northington's review.

The infant most vulnerable to adverse outcome is one born to a polysubstance-abusing mother who does not seek prenatal care and who is nutritionally undernourished. Although methadone treatment programs for opiate-dependent mothers have resulted in improved weight gain and healthier newborns, opiate tolerance and dependence and symptoms of withdrawal can be significant in the newborn infant. Dr. Richardson and colleagues (2006) have focused their review on the effects of prenatal opiate exposure on the developing fetus and the neonatal models that have been used to investigate the mechanisms responsible for opiate tolerance, dependence, and withdrawal. Again, studies in the rodent dominate the scant literature to date. As with the premature infant, the behavioral signs and symptoms of opiate withdrawal are subtle in the newborn rat that has been prenatally exposed to opiates, yet it is quite evident with the aid of molecular probes that opiate exposure affects key neuronal circuits and signaling pathways in the developing brain. Furthermore, the counteradaptive mechanisms associated with opiate exposure in the developing fetus differ from those of the adult. Some of these differences may be related to a changing pattern of the developmental expression of glutamate receptors (specifically the NMDA and AMPA receptors). The rodent model represents a premature model that can be useful in investigating the effect of opiate exposure on the development networks in the brain of premature infants.

Critically ill premature and term infants with respiratory and cardiovascular failure are also frequently exposed to high doses of opiates; thus studies performed in neonatal animal models are useful for investigating the interaction of opiate exposure and the developing brain. Dr. Richardson and colleagues argue that research aimed at understanding the role of opiate exposure, development of tolerance, and dependence will likely lead to effective treatment paradigms for infants with withdrawal syndromes in infancy.

Many of the articles in this issue highlight the use of multiparous rodent models to study biological questions. Although much information has been and will continue to be gained from these models, the nonhuman primate is believed to be the animal model that most accurately approximates the human. Ongoing neonatal studies in baboon models continue to be instrumental in the pathogenesis of respiratory distress syndrome, chronic lung disease (Coalson et al. 1999;Yoder et al. 2000, 2005), and premature brain injury (Inder et al. 2004), although few nonhuman primates have been used in studies of neonatal drug abuse. The nonhuman primate is especially well suited for studying the pathogenesis of perinatal infections because monkeys and humans have similar immunological responses to infection.

Dr. Barry and colleagues (2006) have used the neonatal rhesus macaque model to elucidate the pathogenesis of human congenital cytomegalovirus infection (HCMV¹). Although HCMV infection in the adult and the non-immunocompromised host is of little adverse clinical consequence, HCMV infection is the most common congenital infection in the United States. The developing central nervous system is particularly vulnerable to intrauterine HCMV infection, which can result in permanent neurological damage including mental retardation and sensorineural hearing loss. In developing countries, congenital HCMV infection is the leading nongenetic cause of sensorineuronal hearing loss. The potential for transplacental transmission of HCMV and sequelae in congenitally infected infants is determined by maternal, placental, and fetal factors.

As Dr. Barry and coauthors (2006) discuss, CMV is species specific. Thus even though the rhesus macaque is a robust model for assessing the effects of rhesus macaque CMV (RhCMV¹) infection on the developing fetus, it is not a model of vertical transmission, the mode of transmission in the human fetus. Nevertheless, the macaque system is a model of intrauterine CMV pathogenesis wherein direct injection of RhCMV into the fetal macaque leads to disease similar to that described in the human fetus, and RhCMV can induce severe disease in the fetus that parallels the pathogenesis of HCMV infection in the human fetus. The authors describe exciting findings from these studies, which have demonstrated a predilection of RhCMV to produce cellular changes in the structures of the inner ear.

In contrast to the rhesus macaque model, the guinea pig cytomegalovirus (GPCMV¹) crosses the placenta, causing central nervous system damage and inner ear damage that can result in deafness. Included in this issue is the pertinent review by Dr. Schleiss (2006), who discusses his and colleagues' use of the guinea pig model to study the pathogenesis of congenitally acquired CMV infection. As described, the guinea pig model has been particularly useful in CMV vaccine studies. Dr. Schleiss also discusses the advantages and disadvantages of using other small animal models to study the pathogenesis of CMV infection. Although it is generally accepted that perinatal or postnatally acquired CMV infection in the term newborn does not result in an adverse outcome, this result may not be the case with premature infants. The immunocompromised state of the premature infants makes them vulnerable to infection that increases mortality and morbidity (Bradshaw and Moore 2003; Bryant et al. 2002; Hsu et al. 2001).

Many scientific questions that are investigated in animals arise from epidemiological studies carefully performed in humans. A relative new field, known as developmental origins of adult disease, has demonstrated a strong epidemiological association between undernourished fetuses and the subsequent adverse cardiovascular outcomes during adulthood. The review by Dr. Nathanielsz (2006) highlights how the fetus is remarkably resilient to a suboptimal intrauterine environment (e.g., hypoxia and nutrient restriction) and can adapt to increase the chances of survival, but at a price—an increased risk of morbidity and mortality from cardiovascular disease as an adult. Dr. Nathanielsz presents the 10 fundamental principles of developmental programming in the context of physiological systems involved and the studies in animal models (sheep and rodents) that have been performed to evaluate exposures, mechanisms, and outcomes. While many of these studies are designed to understand the role of the suboptimal intrauterine environment on programming, the neonatal intensive care unit is the "uterine environment" for many premature infants who still have fetal physiology and organs that are following a fetal developmental trajectory that typically results in adverse developmental programming in this infant population.

Many of the articles in this issue discuss ways in which experimentally genetically altered mice can be a powerful tool to equate specific genes with physiological function. However, many disorders are better studied by understanding the interaction of many genes with the environment to produce complex traits. Although it is not highlighted in the reviews in this ILAR Journal issue, there is an emerging awareness that all strains of a particular species may not have an identical expression of many complex traits as fundamental as respiration and cardiovascular function, or vulnerability to infection as pointed out by Dr. Schleiss (2006) in his description of the guinea pig model. The reader is referred to several excellent articles that highlight the fact that strains within a mammalian species may have considerable differences in complex traits (Kacew et al. 1998; Tankersley 2003).

In conclusion, I would like to acknowledge the tireless work of these issue authors in their work with animal models to study diseases of the fetus and newborn. I especially thank the authors for excellent state-of-the-art reviews. It cannot be overstated that well-conducted studies in animal models are relevant and essential to improving the care of the fetus and newborn, particularly the premature infant born at the limit of viability and current understanding of the fetal origins of adult disease. Similar to studies performed in adult models, it is important and necessary to integrate whole animal systems physiology, in vitro cellular biology, and genomic and proteomic approaches in studies of the fetus and newborn.

Finally, I shall end as I began, with the quotation from Dr. Festing's review. Beyond the admonition that "experiments involving neonates should be unbiased, be powerful, have a good range of applicability, not be excessively complex, and be statistically analyzable to show the range of uncertainty in the conclusions," I would like to add that it is imperative that the animal model used be specific to the human condition being studied. This principal is an essential aspect of the responsible use of animals in research.

Acknowledgments

Dr. Gauda acknowledges members of ILAR Council for their helpful advice in conceptualizing this issue, the contributing authors for their excellent reviews, and Susan Vaupel for her exceptional editorial skills and assistance through the publication process.

Abbreviations used in this Introduction: HCMV, human cytomegalovirus; GPCMV, guinea pig cytomegalovirus; RhCMV, rhesus macaque cytomegalovirus.

References

Barry PA, Lockridge KM, Salamat S, Tinling SP, Yue Y, Zhou SS, Gospe SM Jr, Britt WJ, Tarantal AF. 2006. Nonhuman primate models of intrauterine cytomegalovirus infection. ILAR J 47:49-64.

Bradshaw JH, Moore PP. 2003. Perinatal cytomegalovirus infection associated with lung cysts. J Paediatr Child Health 39:563-566.

Bridges JP, Weaver TE. 2006. Use of transgenic mice to study lung morphogenesis and function. ILAR J 47:22-31.

Bryant P, Morley C, Garland S, Curtis N. 2002. Cytomegalovirus transmission from breast milk in premature babies: Does it matter? Arch Dis Child Fetal Neonatal Ed 87:F75-F77.

Coalson JJ, Winter VT, Siler-Khodr T, Yoder BA. 1999. Neonatal chronic lung disease in extremely immature baboons. Am J Respir Crit Care Med 160:1333-1346.

Festing MFW. 2006. Design and statistical methods in studies using animal models of development. ILAR J 47:5-14.

Gaultier C, Matrot B, Gallego J. 2006. Transgenic models to study disorders of respiratory control in newborn mice. ILAR J 47:15-21.

Hsu ML, Cheng SN, Huang CF, Jan CI, Fan HC, Wang CC, Yuh YS. 2001. Perinatal cytomegalovirus infection complicated with pneumonitis and adrenalitis in a premature infant. J Microbiol Immunol Infect 34:297-300.

Inder T, Neil J, Yoder B, Rees S. 2004. Non-human primate models of neonatal brain injury. Semin Perinatol 28:396-404.

Kacew S, Dixit R, Ruben Z. 1998. Diet and rat strain as factors in nervous system function and influence of confounders. Biomed Environ Sci 11:203-217.

Nathanielsz PW. 2006. Animal models that elucidate basic principles of the developmental origins of adult diseases. ILAR J 47:73-82.

Northington FJ. 2006. Brief update on animal models of hypoxic-ischemic encephalopathy and neonatal stroke. ILAR J 47:32-38.

Richardson KA, Yohay A-LJ, Gauda EB, McLemore GL. 2006. Neonatal animal models of opiate withdrawal. ILAR J 47:39-48.

Schleiss MR. 2006. Nonprimate models of congenital cytomegalovirus (CMV) infection: Gaining insight into pathogenesis and prevention of disease in newborns. ILAR J 47:65-72.

Tankersley CG. 2003. Genetic aspects of breathing: On interactions between hypercapnia and hypoxia. Respir Physiol Neurobiol 135:167-178.

Yoder BA, Siler-Khodr T, Winter VT, Coalson JJ. 2000. High-frequency oscillatory ventilation: Effects on lung function, mechanics, and airway cytokines in the immature baboon model for neonatal chronic lung disease. Am J Respir Crit Care Med 162:1867-1876.

Yoder B, Thomson M, Coalson J. 2005. Lung function in immature baboons with respiratory distress syndrome receiving early caffeine therapy: A pilot study. Acta Paediatr 94:92-98.