The disruption of melatonin production in laboratory mouse strains represents
an apparent evolutionary advantage in terms of reproductive development
Animal models can yield valuable insights into the biology of human disorders,
although they can also introduce additional levels of complexity that may make
it a challenge to experimentally untangle the bases for specific phenotypes.
Tadafumi Kato and Takaoki Kasahara of the RIKEN Brain Science Institute in Wako
ran into such a challenge in their attempts to characterize abnormalities in the activity
of melatonin, a hormone that fine-tunes the circadian rhythms that establish an organism’s
day–night cycle, in their mouse model of bipolar disorder. “Since early times, researchers
and psychiatrists have believed that melatonin has something to do with mood disorders,
because many patients experience sleep disturbance and light therapy is used effectively
in the treatment of seasonal affective disorder,” says Kasahara. However, they quickly found
their efforts thwarted by the utter absence of melatonin from their laboratory mice.
Indeed, a growing body of evidence suggests that several strains of laboratory mice—including
the C57BL/6J (B6J) line used by Kato and Kasahara—are deficient in the production of
melatonin, a process that depends on the sequential action of two enzymes: arylalkylamine
N-acetyltransferase (AANAT) and hydroxyindole O-methyltransferase (HIOMT).
Scientists have successfully identified the mouse Aanat gene, and have even uncovered an
inactivating mutation within this gene in B6J mice. However, even with high-quality mouse
genome sequence data available, nobody has yet succeeded in tracking down its partner,
Hiomt. “I studied the mechanism of circadian clocks when I was a PhD student, and I heard
that mouse Hiomt was really enigmatic,” recalls Kasahara, “and even after being away from
the field for about six years, I became aware that mouse Hiomt still had not been identified.”
This is no longer the case, thanks to an extensive analysis of the mouse genome by Kato,
Kasahara and colleagues1. Using the rat HIOMT protein sequence as a basis for comparis
on they have finally managed to uncover this mysterious gene and have thereby revealed
why it has remained hidden from scientists for so long.
Notably, mouse HIOMT bears only limited resemblance to its counterparts in other species,
with an amino acid sequence that is less than 70% identical to that of the rat protein.
Furthermore, this gene was likely masked by its residence within the pseudoautosomal
region (PAR), a poorly characterized stretch of DNA within the sex chromosomes that
enables them to efficiently ‘pair up’ and undergo recombination during meiosis. “The
PAR contains extremely repetitive sequences and high guanine-cytosine content, both
of which make it difficult to sequence using either traditional or next-generation sequencing
methods,” says Kasahara.
Closer analysis of the sequence of this gene revealed two notable sequence variations in
B6J mice relative to MSM animals—a strain derived more recently from wild mice that exhibits
normal melatonin production. Both of these changes affect the amino acid sequence of the
encoded protein, and the investigators showed that each mutation leads to a strong reduction
in HIOMT levels. These mutations also proved to be widespread among a variety of other
inbred mouse strains, including several lines commonly employed in laboratory research.
Kato, Kasahara and colleagues also noted that although this HIOMT deficiency appears to
have a limited impact on circadian behaviors, it has a clear effect on gonadal development;
melatonin-deficient animals with the B6J versions of the Hiomt and/or Aanat genes exhibited
significantly greater testicular growth than their melatonin-producing counterparts. Conversely,
in experiments with ICR mice, another melatonin-deficient strain, the researchers showed that
treatment with melatonin was associated with a reduction in testicular weight.
These findings are in keeping with other data showing a vital link between melatonin and
reproductive development—including observations in human patients. “Children with little
or no melatonin due to pineal tumors often show premature sexual maturation,” says
Evolution in a cage
Because these defects appear to be specifically prevalent among cultivated strains of
laboratory mice, it appears likely that there is some manner of selection taking place that
favors the emergence of strains with reduced melatonin levels and accelerated
reproductive development—even if this evolution was unintentional and, until now,
invisible. This finding is supported by similar research in domesticated chickens,
which has spotlighted the emergence of other gene variations that may potentially
influence the same developmental pathway2. “One of the most intriguing [variants] is
found in the gene encoding the receptor for thyroid-stimulating hormone, because
TSH and melatonin are closely related in seasonal breeding,” says Kasahara.
These findings could also have potential implications for previous animal studies
that have investigated circadian rhythms, given that much of this research has been
conducted in B6J and other inbred strains. For example, one recent study has shown
that the circadian rhythm defects observed in the widely used B6J-derived Clock mutant
mice are markedly diminished in the presence of normal levels of melatonin.
These findings will closely inform future work from Kasahara and Kato, who are in the process
of engineering a melatonin-producing B6J strain for use in their future investigations of
mood disorders. However, Kasahara also suggests that conventional laboratory strains
in general may be too interbred for their own good. “Our B6J mouse model for mood
disorder has many phenotypes similar to bipolar disorder, but they don’t get manic
spontaneously,” he says. “I hypothesize that laboratory mice have lost their potential
to develop manic or aggressive episodes, and we are consequently using wild-derived
mice, which are very aggressive, alert and agile in order to study these disorders.”