Science - USA (2021-07-09)

SCIENCE sciencemag.org 9 JULY 2021 • VOL 373 ISSUE 6551 155

be described as a key innovation—allowed

hummingbirds to exploit nectar, a resource

that most early birds, not only the ancestor

of hummingbirds, most likely could not taste.

This ancestral absence of sweet perception

can be inferred because in most nonavian

lineages, the sweet receptor is formed by a

heterodimer of two taste receptors, T1R2 and

T1R3, the former of which is missing in birds

whose genomes have been sequenced (now

spanning the entire avian tree) ( 10 ). The

umami receptor is formed by pairing T1R3

with another molecular partner, T1R1, which

is still present in bird genomes. Toda et al.

suggest that the evolution of the umami re-

ceptor may have played a critical permissive

role in some of the classic examples of avian

adaptive radiation (e.g., the Darwin’s finches

and honeycreepers) as well.

To d a et al. analyzed taste receptor func-

tion and evolution in passerine birds.

Passerines are members of the largest of

40 orders of birds, which comprises more

than half of all bird species. The authors re-

viewed the frequency of nectar consumption

(the most common source of dietary sugars)

across birds, identifying multiple nonpas-

serine (hummingbirds, parrots, and others)

and many passerine lineages that extensively

consume sugars. Members of the oscine pas-

serines—commonly known as songbirds—

including well-known nectarivore radia-

tions, such as the honeyeaters, sunbirds,

and Hawaiian honeycreepers, as well as

less-specialized groups, such as wattlebirds,

white-eyes, and tanagers, frequently imbibe

nectar and similar sugar sources, such as

honeydew. Choice tests in a nectar special-

ist (a honeyeater) and a nonspecialist (the

canary) demonstrate that oscine passerines

both consume and taste sugars. A compara-

tive evolutionary analysis of the pattern of

nectarivory across birds reported by Toda

et al. suggests that perception of sweetness

may be hard to acquire—with possibly a sin-

gle origin within oscines—but, once present,

this trait permits rapid gains and losses of

sugar exploitation.

To interrogate the molecular basis of

sweet perception, Toda et al. cloned the T1R

genes from species spanning most of the pas-

serine evolutionary tree. This allowed them

to express these native genes (as well as het-

erospecific combinations and mutants) in

cell cultures and detect ligand binding using

a cell-based luminescence assay. All tested

oscine umami receptors (T1R1-T1R3 dimers)

responded strongly to sugars, especially su-

crose. By contrast, umami receptors from

the two nonoscine passerines retained the

ancestral umami sensitivity with no sign of

activation by sugar. Expression studies com-

bining honeyeater T1R receptor subunits

with the subunits of other oscine species

demonstrate conserved sweet perception—

and thus presumably a shared molecular

mechanism—across most oscines. By con-

trast, coexpression of these genes with their

partners cloned from hummingbirds showed

no binding activity for sugars, though they

still responded to amino acids, which sug-

gests a different binding mechanism be-

tween these deeply divergent lineages.

The shared oscine binding mechanism

was further explored by synthesizing in-

ferred ancestral genes for T1R1 and T1R3 and

expressing these and chimeras of the two to

test for sugar sensitivity. These experiments

suggest that the origin of shared sweet per-

ception is nested somewhat within the os-

cine lineage (excluding two major branches

of oscines) and that it involves evolution of

residues in both T1R1 and T1R3 subunits,

whereas previous work on hummingbirds

identified most changes in T1R3. Functional

convergence in these lineages is therefore

likely based on complex changes in tertiary

structure evolving from different starting

points, rather than parallel residue-for-resi-

due replacements, which might explain the

relatively few origins of this trait.

The diversity of passerines has long in-

trigued biologists ( 11 ), and some oscine lin-

eages containing nectarivores have unusu-

ally high species diversity given their age ( 8 ,

12 , 13 ). Biogeographic analyses of oscines in-

dicate that they originated in Australasia ( 13 ,

14 , 15 ), and thus sweet perception probably

evolved there in the Oligocene (34 to 23 mil-

lion years ago) or possibly earlier. The early

evolution of sweet perception likely played

an important role in diversification of this

lineage, which is now a numerically and eco-

logically dominant component of terrestrial

avifaunas the world over. j

REFERENCES AND NOTES

1. Y. Toda et al., Science 373 , 226 (2021).


2. Z. Liu, F.-Y. Qi, D.-M. Xu, X. Zhou, P. Shi, Sci. Adv. 4 ,
   eaat8821 (2018).


3. R. Borges et al., BMC Genomics 16 , 751 (2015).


4. L. R. Yohe et al., Evolution 71 , 923 (2017).


5. M. W. Baldwin et al., Science 345 , 929 (2014).


6. L. S. Carvalho, D. M. A. Pessoa, J. K. Mountford, W. I. L.
   Davies, D. M. Hunt, Front. Ecol. Evol. 5 , 34 (2017).


7. A. A. Fernandez, M. R. Morris, Am. Nat. 170 , 10 (2007).


8. W. Jetz, G. H. Thomas, J. B. Joy, K. Hartmann, A. O.
   Mooers, Nature 491 , 444 (2012).


9. J. A. McGuire et al., Curr. Biol. 24 , 910 (2014).


10. S. Fe n g et al., Nature 587 , 252 (2020).


11. R. J. Raikow, Syst. Zool. 35 , 255 (1986).


12. R. G. Moyle, C. E. Filardi, C. E. Smith, J. Diamond, Proc.
    Natl. Acad. Sci. U.S.A. 106 , 1863 (2009).


13. C. H. Oliveros et al., Proc. Natl. Acad. Sci. U.S.A. 116 , 7916
    (2019).


14. R. G. Moyle et al., Nat. Commun. 7 , 12709 (2016).


15. F. K. Barker, A. Cibois, P. Schikler, J. Feinstein, J. Cracraft,
    Proc. Natl. Acad. Sci. U.S.A. 101 , 11040 (2004).
    10.1126/science.abj6746

CANCER

Sowing

the seeds of

leukemia

before birth

By Irene Roberts1,2 and Paresh Vyas^1

E

ach year, ~200,000 babies worldwide

are born with Down syndrome (DS),

owing to constitutional trisomy of

chromosome 21 (T21) ( 1 ). Children with

DS have a markedly increased risk of

leukemia, particularly in their first

4 years. Almost 60,000 (30%) will harbor

within their blood cells damaging, fetally ac-

quired mutations in the transcription factor

gene GATA binding protein 1 (GATA1), which

encodes a short GATA1 protein (GATA1s)

and triggers the first step in the develop-

ment of leukemia ( 2 ). GATA1 mutations are

rare in disomic individuals and virtually

never cause leukemia in the absence of T21.

Why GATA1 mutations are so frequent in

T21 babies and the mechanisms by which

a supernumerary chromosome 21 (Hsa21)

predisposes to, and cooperates with, genetic

events in DS leukemogenesis are not known.

On page 179 of this issue, Wagenblast et al.

( 3 ) identify Hsa21 microRNAs (miRNAs) that

cooperate with GATA1s and map the cellular

origin of the leukemia.

Compared with children of the same age

without DS, the risk of myeloid leukemia

(modeled by Wagenblast et al.) is 150-fold

greater, whereas acute lymphoblastic leuke-

mia is more than 20-fold higher in people

with DS. DS also causes functional and de-

velopmental abnormalities in virtually all

organs and tissues. Progressive cognitive

impairment, dementia, and cardiac disease

are major causes of ill health, with profound

effects on the lives of patients and their fami-

lies. How the supernumerary Hsa21 causes

the clinical features of DS remains unclear.

Distinct cooperation

between a mutated gene

and trisomy 21 triggers

leukemia in utero

(^1) MRC Molecular Haematology Unit, MRC Weatherall
Institute of Molecular Medicine, University of Oxford, Oxford,
UK.^2 Department of Paediatrics, University of Oxford,
Oxford, UK. Email: [email protected];
[email protected]
The early evolution of sweet perception
likely played an important role in the
diversification of passerines, such as this
New Holland honeyeater in Australia.
0709Perspectives.indd 155 7/1/21 6:34 PM