Sudden infant death syndrome and serotonin:
Dartmouth Medical School, Lebanon, NH, USA
The sudden infant death syndrome (SIDS) is the sudden,
unexpected death of an infant that is not explained by
autopsy, death scene examination, and history. The
etiology is unknown. Recent postmortem studies have
discovered abnormalities in brainstem serotonergic neu-
rons, but how these translate into dysfunction and cause
SIDS is uncertain. Recently, lethal effects in transgenic
mice with overexpression of the serotonin 1A autorecep-
tor have been described. Many die spontaneously
between postnatal day 40 (P40) and P80, and some
spontaneously exhibit bradycardias and drops in body
temperature. The severity of the autonomic dysfunction
and its age dependence suggest relevance to SIDS.
However, SIDS cases have decreased serotonin 1A auto-
in the mice, and the peak incidence of SIDS is between 2
and 6 months of age, which is arguably younger (in
relative terms) than the ages at which the mice die.
Nevertheless, the description of an animal model with
serotonin defects that has autonomic dysfunction and
spontaneous mortality at a young age is an exciting
finding of possible importance for understanding SIDS.
Keywords: 5-HT; brainstem; SIDS
SIDS is the sudden, unexpected death of an infant that is not
explained by autopsy, death scene examination, and
history.(1)Overall there are currently ?2000 SIDS deaths
per year intheUS andall motherscertainly worryaboutSIDS.
There are many theories to explain SIDS, but recent work has
focused attentionon serotonergic neurons inthe brainstem.(2)
Postmortem examination of brains from SIDS cases has
revealed, in a majority of cases, a triad of serotonergic
abnormalities(2)including (i) increased number of serotonin
neurons many of which appear immature, (ii) decreased
binding of the serotonin 1A autoreceptor, and (iii) decreased
relative serotonin transporter binding. Further, four genetic
studies have shown a higher incidence in SIDScases of the L/
L genotype and/or the long (L) allele in the promotor region for
the serotonin transporter gene,(3–6)
although a similar
examination of the coding region for the serotonin 1A
autoreceptor did not demonstrate any link with SIDS.(7)
These data point to serotonin abnormalities in many SIDS
cases, but are difficult to interpret in any mechanistic way.
Does SIDS have a genetic origin? Or is it the result of an
environmental stress? Or both? Are the serotonergic
abnormalities a primary event or an epiphenomenon? How
might the serotonergic abnormalities result in physiological
dysfunction that contributes to SIDS?.
Triple risk hypothesis
The triple risk hypothesis has proven useful in thinking about
SIDS causation.(8)It posits that SIDS results from the
unfortunate coincidence of (i) an infant with specific under-
homeostatic stressor at the time of death. The serotonin
abnormalities could contribute to the underlying vulnerability.
SIDS would then require a stressor that occurs at the critical
period, which is between 2 and 6 months of age..
Animal models that alter function of serotonin neurons during
early postnatal life are useful in understanding how the
serotonergic system affects autonomic and respiratory
control processes that could contribute to sudden death.
They provide insight into possible mechanisms for SIDS.
These have taken two generic forms: pharmacological
manipulation in early postnatal life; and alteration of a single
gene function from inception. In the first, the dysfunction
begins in early postnatal life; in the second it is present from
early prenatal life.
Loss or inhibition of serotonergic
Studieswith lesions of serotonergic neurons in the medulla by
early postnatal injection of toxins via the cisterna magna in
newborn piglets have shown that the ventilatory response to
increased CO2is reduced in males during sleep (there is a
male predominance in the incidence of SIDS).(9)Focal acute
What the papers sayDOI 10.1002/bies.200800200
*Correspondence to: E. Nattie, Borwell Bldg, Dartmouth Medical School,
Lebanon, NH 03756-0001, USA
BioEssays 31:130–133, ? 2009 WILEYPeriodicals,Inc.
inhibition of serotonergic neurons in the medullary raphe of
newborn piglets by reverse dialysis of a serotonin 1A
autoreceptor agonist decreases the ventilatory response
to CO2,(10)disrupts sleep,(11)and alters thermoregulation and
the ability to regulate heart rate in cold stress.(12)Focal
inhibition of the medullary raphe by a GABA-A receptor
agonist also decreases the ventilatory response to CO2(13)
and prolongs the apnea induced by placing water in the larynx
(the laryngeal chemoreflex).(14)Disruption of the medullary
raphe in early postnatal life can alter the function of reflexes
that sustain viability.
Adult transgenic mice with conditional loss of all seroto-
nergic neurons have a reduced ventilatory response to CO2
and an inability to maintain body temperature in cold
stress,(15)even as mild as exposure to 258C.(16)The
ventilatory response to hypoxia appears to be intact. Adult
transgenic mice with absence of the serotonin transport
protein show a dramatic decrease in the ventilatory response
to CO2, but in males only.(17)The hypoxic response appears
intact. It will be of interest to see the autonomic and
respiratory control function of both of these mice in early
Transgenic mice with loss of a transcription factor, pet-1,
which determines the fate of developing serotonergic
neurons, exhibit a decrease in the numbers of serotonin
them.(18)These mice show distinct abnormalities in breathing
but only in early postnatal life; subsequently, they appear to
recover normal breathing.(19)However, some of these mice
die and if these mice are exposed to another stress, e.g.,
pathogens in the environment, their abnormalities are more
severe and their death is more likely.(19)
Overexpression of the serotonin 1A
Transgenic mice with an overexpression of the serotonin 1A
autoreceptor under the control of a tetracycline activator, bred
to study behavior, have been found, serendipitously, to die
suddenly and unexpectedly in large numbers.(20)In these
mice the serotonin 1A autoreceptor overexpression can be
reversed by administration of doxycycline. Mice exhibiting
overexpression of the serotonin 1A autoreceptor in the
medulla from early prenatal life begin to die at postnatal
day 25 (P25) with most deaths occurring between P40 and
P80. There is ?40% survival. If the overexpression of
serotonin 1A autoreceptor begins at P40, again there is
mortality with only ?10% survival to adulthood. If the onset of
the overexpression is delayed until P60, mortality is less.
How do these mice die? Study of a subgroup of mice
with onset of the serotonin 1A autoreceptor overexpression
at P60 (those with lower mortality) showed spontaneous
bradycardias that preceded drops in body temperature.
In some, death occurred during one of these spontaneous
episodes. Thesemicewere also
brown adipose tissue to maintain normal body temperature
when exposed to cold (48C).
These dramatic observations underscore the potential
importance of the serotonergic system in the maintenance of
normal autonomic function. They complement the findings
from other studies outlined above. Are they relevant to SIDS?
One obvious concern is that SIDS cases have a decreased
binding for the serotonin 1A autoreceptor, while these mice
have an overexpression. A second is the age of the mortality.
The peak incidence of SIDS is between 2 and 6 months
postnatal age. What is the equivalent age in developing mice?
Audero et al.(20)argue from a comparative whole brain
anatomy analysis(21)that P40 may be equivalent to early
human postnatal life. This seems untenable for autonomic
and physiological function. Mice open their eyes and begin to
hear at ?P12, they are weaned at ?P25 and are
reproductively capable at P40. Analysis of physiological
function attributable to the brainstem, e.g., the ventilatory
response to CO2, suggests three periods: early development,
P0–P10, like premature infants; transitional phase, P11–P14,
infants at or just after birth; and late development, P15–P21,
similar to early postnatal life.(22)
unable to activate
Conclusions from the animal models
Is there any unifying theme with respect to SIDS causation
that arises from the transgenic mouse studies? Mice with few
or absent serotonergic neurons have low or absent extra-
cellular serotonin levels. Mice with absent serotonin trans-
porter function would have increased extracellular serotonin
with overexpression of the serotonin 1A autoreceptor would
arguably have decreased extracellular serotonin levels,
although we do not know this as yet. Extracellular levels of
serotonin do not appear to be a common attribute among
these transgenic mice models.
Could it be the ‘‘gain’’ or ‘‘responsiveness’’ of the
serotonergic system? The mice with few or absent seroto-
nergic neurons would have decreased ‘‘responsiveness’’, as
would the mice with overexpressed serotonin 1A autorecep-
the serotonergic system also support this idea; inhibition or
lesions of the medullary raphe decrease function, as
measured by breathing reflexes,sleep, and thermoregulation.
What about the serotonin transporter-deficient mice? In this
case the net functional result will likely depend on the balance
of effects due to: (i) the chronically increased extracellular
serotonin, which would make the system less ‘‘responsive’’
due to chronic autoinhibition, and (ii) any long-term adaptive
E. Nattie What the papers say
BioEssays 31:130–133, ? 2009 WILEYPeriodicals,Inc.
decrease in the serotonin 1A autoreceptor,(17)which would
minimize the degree of chronic autoinhibition. That a
decreasedventilatoryresponse toCO2hasbeen observed(17)
supports a less ‘‘responsive’’ system.
How would a less ‘‘responsive’’ medullary serotonergic
system result in SIDS? If this system acts to provide an
excitatory modulation of autonomic and respiratory control,
and this modulation is especially important in situations of
additional stress, a less ‘‘responsive’’ system could be
detrimental. For example, in the studies of the pet-1-knockout
mice by Erickson et al.,(19)the mice living in an environment
with pathogens faired less well, although mice in both
environments were transiently abnormal. The mice with
overexpression of the serotonin 1A autoreceptor appeared to
have episodes of bradycardia and to die spontaneously,
without any applied stress. Here we do not know what stress
might have caused the onset of these episodes. The authors
suggest a link with sudden changes in sleep state, an
important possibility as SIDS occurs in sleep.
Implications for the causes of SIDS
Studies in transgenic and pharmacologically manipulated
animal models suggest that a less responsive serotonergic
system could be detrimental to physiological control systems
and possibly contribute to unexpected death. How does this
SIDS cases indicates increased numbers of serotonin
neurons, decreased 1A autoreceptor binding, and decreased
transporter binding per neuron, which could be interpreted as
increased or decreased ‘‘responsivity’’ depending on the
functional balance of the defects. Less autoinhibition would
suggest greater ‘‘responsivity’’ but decreased re-uptake with
more synaptic serotonin would suggest decreased ‘‘respon-
sivity’’. We do not know which outcome is present in the SIDS
cases. The gene-association studies link SIDS to both the L
allele and the L/L genotype for the serotonin transporter,
which predicts increased transporter function, not the
decreased binding as observed in SIDS cases.(2)This should
lower extracellular serotonin levels, which should induce an
adaptive increase in serotonin 1A autoreceptor binding, a
prediction opposite to what is observed in SIDS cases.(2)
Further complicating this analysis is the role of serotonin in
development; early genetic alteration of the serotonin system
could result in abnormalities and dysfunction of non-
serotonergic neurons. While the ongoing studies in animal
models point to a significant role for medullary serotonergic
we do not know as yet how to relate the specific anatomical
abnormalities described in SIDS cases to specific functional
deficits attributable to the serotonergic system.
1. Willinger,M.,James,L.S.andCatz,C.,Definingthesudden infantdeath
syndrome (SIDS): deliberations of an expert panel convened by the
National Institute of Child Health and Human Development. Pediatr Pathol
1991. 11: 677–684.
2. Paterson, D. S., Trachtenberg, F. L., Thompson, E. G., Belliveau,
R.A., Beggs,A. H., etal.
abnormalities in sudden infant death syndrome. JAMA 2006. 296:
3. Narita, N., Narita, M., Takashima, S., Nakayama, M., Nagai, T. and
Ohado, N., Serotonin transporter gene variation is a risk factor for sudden
infant death syndrome in the Japanese population. Pediatrics 2001. 107:
4. Weese-Mayer, D. E., Berry-Kravitz, E. M., Maher, B. S., Silvestri, J. M.,
Curran, M. E. and Marazita, M. L., Sudden infant death syndrome:
association with a promotor polymorphism of the serotonin transporter
gene. Am J Genet Part A 2003. 117: 268–274.
5. Opdal, S. H., Vege, A˚. and Rognum, T. O., Serotonin transporter gene
variation in sudden infant death syndrome. Acta Paediatr 2008. 97: 861–
6. Marzano, F. N., Maldini, M., Filzoni, L., Lavezzi, A. M., Parmigiani, S.,
etal.Geneticvariation inserotonin metabolismand SIDS.Genomics2008.
7. Morley, M. E., Rand, C. M., Berry-Kravitz, E. M., Zhou, L., Fan, W. and
Weese-Mayer, D. E., Genetic variation in the HTR1A gene and sudden
infant death syndrome. Am J Genet Part A 2008. 146: 930–933.
8. Filiano, J. J. and Kinney, H. C., A perspective on neuropathologic
findings in victims of the sudden infant death syndrome: the triple-risk
model. Biol Neonate 1994. 65: 194–197.
9. Penatti, E. M., Berniker, A. V., Kereshi, B., Cafaro, C., Kelly, M. L., et al.
Ventilatory response to hypercapnia and hypoxia after extensive lesion of
medullary serotonergic neurons in newborn conscious piglets. J Appl
Physiol 2006. 101: 1177–1188.
10. Messier, M. L., Li, A. and Nattie, E. E., Inhibition of medullary raphe
serotonergic neurons has age-dependent effects on the CO2response in
newborn piglets. J Appl Physiol 2004. 96: 1909–1919.
11. Messier, M. L., Li, A. and Nattie, E. E., Muscimol inhibition of medullary
raphe ´ neurons decreases the CO2response and alters sleep in newborn
piglets. Respir Physiol Neurobiol 2002. 133: 197–214.
12. Brown, J. W., Sirlin, E. A., Benoit, A. M., Hoffman, J. M. and Darnall,
R. A., Activation of 5-HT1A receptors in medullary raphe disrupts sleep
and decreases shivering during cooling in the conscious piglet. Am J
Physiol Regul Integr Comp Physiol 2008. 294: R884–R894.
13. Curran, A. K., Darnall, R. A., Filiano, J. J., Li, A. and Nattie, E. E.,
Muscimol dialysis in the rostral ventral medulla reduced the CO(2)
response in awake and sleeping piglets. J Appl Physiol 2001. 90: 971–
14. Van Der Velde, L., Curran, A. K., Filiano, J. J., Darnall, R. A., Bartlett,
D., Jr. and Leiter, J. C., Prolongation of the laryngeal chemoreflex after
inhibition of the rostral ventral medulla in piglets: a role in SIDS? J Appl
Physiol 2003. 94: 1883–1895.
15. Hodges, M. R., Tattersall, G. T., Harris, M. B., McEvoy, S. D., Richer-
son, D. N., et al. Defects in breathing and thermoregulation in mice with
near-completeabsenceofcentral serotonin neurons.JNeurosci 2008.28:
16. Hodges, M. R. and Richerson, G. B., Interaction between defects in
ventilatory and thermoregulatory control in mice lacking 5-HT neurons.
Respir Physiol Neurobiol 2008. P [Epub ahead of print].
17. Li, A. and Nattie, E., Serotonin transporter knockout mice have a reduced
ventilatory response to hypercapnia (predominantly in males) but not to
hypoxia. J Physiol 2008. 586: 2321–2329.
18. Hendricks, T. J., Fyodorov, D. V., Wegman, L. J., Lelutiu, N. B., Pehek,
E. A., et al. Pet-1 ETS gene plays a critical role in 5-HT neuron develop-
ment and is required for normal anxiety-like and aggressive behavior.
Neuron 2003. 37: 233–247.
19. Erickson, J. T., Shafer, G., Rossetti, M. D., Wilson, C. G. and Deneris,
E. S., Arrest of 5HT neuron differentiation delays respiratory maturation
What the papers sayE. Nattie
BioEssays 31:130–133, ? 2009 WILEYPeriodicals,Inc.
and impairs neonatal homeostatic responses to environmental chal-
lenges. Respir Physiol Neurobiol 2007. 159: 85–101.
20. Audero, E., Coppi, E., Mlinar, B., Rossetti, T., Caprioli, A., et al.
Sporadic autonomic dysregulation and death associated with excessive
serotonin autoinhibition. Science 2008. 321: 130–133.
21. Clancy, B., Darlington, R. B. and Finlay, B. L., Translating develop-
mental time across mammalian species. Neuroscience 2001. 105: 7–17.
22. Davis, S. E., Solhied, G., Castillo, M., Dwinnell, M., Brozoski, D. and
Forster, H. V., Postnatal developmental changes in CO2sensitivity in rats.
J Appl Physiol 2006. 101: 1097–1103.
E. NattieWhat the papers say
BioEssays 31:130–133, ? 2009 WILEYPeriodicals,Inc.