ArticlePDF Available

Human GNPTAB stuttering mutations engineered into mice cause vocalization deficits and astrocyte pathology in the corpus callosum


Abstract and Figures

Significance Stuttering is a common neurodevelopmental disorder. However, the neurological causes of this disorder are poorly understood. The disorder is highly genetic, and recent discoveries have found several genes involved in this disorder, but how these lead to the unique clinical features of stuttering has been unknown. We have shown that mice carrying human stuttering mutations display vocalization deficits that recreate the salient features of human stuttering. In these mice, we have used several different complementary techniques to identify a specific deficit in astrocytes, a type of glial cell prominent in white matter tracts, particularly in the corpus callosum of these mice. These findings suggest astrocytes as a site of the neuropathology, leading to a deficit in interhemispheric connectivity in stuttering.
Increased pause durations in the vocalizations of Gnptab Ser321Gly (S321G) and Ala455Ser (A455S) compared to wild-type littermates. All graphs indicate pause lengths in vocalizations of individual animals (1 point = 1 animal). The average of individual mean pause lengths in each genotype group was compared by t test calculating 2-tailed P value. Green represents wild-type ( +/+ ), blue represents heterozygous knockin (+/mut), red represents homozygous knockin (mut/mut). All P values depicted in the figures are comparison between +/+ and mut/mut. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001. Error bars indicate the SEM. (A) Interbout pause duration analysis in S321G animals displaying n > 15 syllables. Sample sizes (n) of +/+ = 46, n of +/mut = 75, n of mut/mut = 40. (B) Interbout pause duration analysis with animals displaying total n > 50 syllables. The n of +/+ = 45, n of +/mut = 71, n of mut/mut = 32. (C) Interbout pause duration analysis in A455S animals displaying n > 15 syllables. The n of +/+ = 31, n of +/mut = 63, n of mut/mut = 30. (D) Interbout pause duration analysis in A455S animals displaying n > 50 syllables. The n of +/+ = 29, n of +/mut = 62, n of mut/mut = 30. (E) Intrabout pause duration analysis in S321G animals displaying total n > 15 syllables. The number of animals of each genotype are the same as that in A. (F) Intrabout pause duration analysis in S321G animals displaying total n > 50 syllables. The number of animals of each genotype are the same as that in B. (G) Intrabout pause duration analysis in A455S animals displaying total n > 15 syllables. The number of animals of each genotype are the same as that in C. (H) Intrabout pause duration analysis in A455S animals displaying total n > 15 syllables. The number of animals of each genotype are the same as that in D. Degrees of freedom (DF) in A and E is 84, in B and F is 75, in C and G is 59, in D and H is 57.
Content may be subject to copyright.
Human GNPTAB stuttering mutations engineered into
mice cause vocalization deficits and astrocyte
pathology in the corpus callosum
Tae-Un Han
, Jessica Root
, Laura D. Reyes
, Elizabeth B. Huchinson
, Johann du Hoffmann
, Wang-Sik Lee
Terra D. Barnes
, and Dennis Drayna
Section on Genetics of Communication Disorders, National Institute on Deafness and Other Communication Disorders, National Institutes of Health,
Bethesda, MD 20892;
Section on Quantitative Medical Imaging, National Institute of Biomedical Imaging and Bioengineering, National Institutes of
Health, Bethesda, MD 20892;
Section on Behavioral Neuroscience, Rodent Behavioral Core, National Institute of Mental Health, National Institutes of
Health, Bethesda, MD 20892;
Department of Internal Medicine, Washington University School of Medicine in St. Louis, St. Louis, MO 63110;
Department of Neuroscience, Washington University School of Medicine in St. Louis, St. Louis, MO 63110
Edited by Stephen T. Warren, Emory University School of Medicine, Atlanta, GA, and approved July 18, 2019 (received for review January 28, 2019)
Stuttering is a common neurodevelopmental disorder that has
been associated with mutations in genes involved in intracellular
trafficking. However, the cellular mechanisms leading to stuttering
remain unknown. Engineering a mutation in N-acetylglucosamine-1-
phosphate transferase subunits αand β(GNPTAB) found in humans
who stutter into the mouse Gnptab gene resulted in deficits in the
flow of ultrasonic vocalizations similar to speech deficits of humans
who stutter. Here we show that other human stuttering mutations
introduced into this mouse gene, Gnptab Ser321Gly and Ala455Ser,
produce the same vocalization deficit in 8-day-old pup isolation calls
and do not affect other nonvocal behaviors. Immunohistochemistry
showed a marked decrease in staining of astrocytes, particularly
in the corpus callosum of the Gnptab Ser321Gly homozygote mice
compared to wild-type littermates, while the staining of cerebellar
Purkinje cells, oligodendrocytes, microglial cells, and dopaminergic
neurons was not significantly different. Diffusion tensor imaging
also detected deficits in the corpus callosum of the Gnptab Ser321Gly
mice. Using a range of cell type-specific Cre-drivers and a Gnptab
conditional knockout line, we found that only astrocyte-specific
Gnptab-deficient mice displayed a similar vocalization deficit. These
data suggest that vocalization defects in mice carrying human stut-
tering mutations in Gnptab derive from abnormalities in astrocytes,
particularly in the corpus callosum, and provide support for hypoth-
eses that focus on deficits in interhemispheric communication in
white matter
mouse vocalization
Stuttering is a common neurodevelopmental disorder charac-
terized by disruptions in the fluent flow of speech (1), typi-
cally in the absence of other neurological deficits. Stuttering
displays high heritability (2), and recent studies have identified
mutations in the GNPTAB,GNPTG,NAGPA, and AP4E1 genes
that are associated with this disorder (35). The products of these
genes interact with each other in vivo and in vitro (3, 4), and
participate in the control of intracellular trafficking, deficits in
which are recognized in a wide range of neurological disorders (6).
A primary goal for stuttering research has been to understand
the neuropathology underlying this disorder. Imaging studies
have identified differences in the brains of individuals who
stutter (710). However, such studies have been limited by the
difficulty of determining whether these differences are the cause
of stuttering or the result of stuttering, and by the fact that they
do not provide resolution at the cellular and molecular level. The
identification of mutations in specific genes associated with hu-
man stuttering has allowed the construction of mouse models of
the disorder. Mice display rich, context-specific ultrasonic vocali-
zations (USVs) that have become increasingly well characterized
(1114) and have been found to be under substantial genetic
control (1517). In addition, the brain anatomy and circuitry for
vocalization in the mouse has been shown to share similarities with
those of humans (18). In addition, mice carrying mutations in
FoxP2, a gene mutated in human developmental verbal dyspraxia,
have been shown to have abnormalities in a range of vocalization
phenotypes (19, 20), supporting the view that mice can serve as a
valid model for investigating central nervous system functions
associated with the control of vocalization. Aided by our increasing
understanding of mouse vocalization, we have generated a mouse
model of stuttering by engineering a common stuttering mutation
in the Gnptab gene encoding N-acetylglucosamine-1-phosphate
transferase subunits αand β(21) into the mouse germline.
While the ultrasonic pup isolation calls of such mice display many
normal features, they have abnormally long pauses in their stream
of vocalization compared to their wild-type littermates, and these
abnormal pauses are similar to those observed in the speech of the
humans who carry such mutations (22).
In this study, we sought to confirm and expand our under-
standing of the effects of GNPTAB mutations found in human
stuttering on mouse USV, and to use such mutant mice to in-
vestigate the neuropathology present in these animals. To do this,
we introduced additional human stuttering mutations in Gnptab
Stuttering is a common neurodevelopmental disorder. How-
ever, the neurological causes of this disorder are poorly un-
derstood. The disorder is highly genetic, and recent discoveries
have found several genes involved in this disorder, but how
these lead to the unique clinical features of stuttering has been
unknown. We have shown that mice carrying human stuttering
mutations display vocalization deficits that recreate the salient
features of human stuttering. In these mice, we have used
several different complementary techniques to identify a spe-
cific deficit in astrocytes, a type of glial cell prominent in white
matter tracts, particularly in the corpus callosum of these mice.
These findings suggest astrocytes as a site of the neuropa-
thology, leading to a deficit in interhemispheric connectivity
in stuttering.
Author contributions: T.-U.H., J .d.H., and D.D. designe d research; T.-U.H., J.R ., L.D.R.,
E.B.H., and W.-S.L. performed research; T.-U.H. contributed new reagents/analytic tools;
T.-U.H., J.R., E.B.H., J.d.H., W.-S.L., T.D.B., and D.D. analyzed data; and T.-U.H., J.d.H., and
D.D. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
Published under the PNAS license.
To whom correspondence may be addressed. Email:
This article contains supporting information online at
1073/pnas.1901480116/-/DCSupplemental. PNAS Latest Articles
Downloaded by guest on October 7, 2020
into mice and tested the USVs in these mice. We performed
immunohistochemistry studies of the brains of these animals, fo-
cusing on cell types and brain regions previously suggested to be
involved in human stuttering. To obtain orthogonal evidence re-
garding the role of brain regions and cell types in the disordered
vocalizations of mice carrying human stuttering mutations, we
selectively knocked out the Gnptab gene in specific brain cell types
or lineages using a range of neuronal Cre-driver mouse strains (23,
24), and tested the USVs in these animals.
Abnormal USVs in Knockin Mice Carrying Mutations Found in Human
Stuttering. A previous study of mice carrying the human stut-
tering mutation GNPTAB Glu1200Lys in the orthologous mouse
Gnptab gene showed that vocalization anomalies of mutant mice
were most pronounced at postnatal days 3 to 8 (P3 to P8) (22).
We recorded these pup isolation calls at P8 and analyzed their
rates, spectral characteristics, and timing. Barnes et al. (22) found
differences in the timing of vocalizations between homozygote
mutant (mut/mut) and their wild-type littermates, with increased
pauses between bouts of syllables in mutant mice and abnormally
long pauses in the stream of their vocalization that are similar to
those detected using a quantitative analysis of the vocalizations of
humans who stutter carrying such mutations (22). In addition, the
entropy of their vocalization temporal sequencing was reduced,
consistent with increased stereotypy in their vocalizations.
To confirm and extend these results, we constructed mouse
lines carrying other mutations found in human stuttering, in-
cluding the Ser321Gly and Ala455Ser mutations in Gnptab
(constructions detailed in SI Appendix,SI Materials and Methods
and Fig. S1). With these mice, we first focused on analyses of
pause durations in their USVs at P8. We calculated the mean
pause lengths in each individual animal and compared the average
of the individual mean pause lengths in each genotype group.
Pause lengths were classified into 2 types, interbout and
intrabout pauses (SI Appendix, Fig. S2A). Interbout pauses are
long pauses between groups of consecutive syllables (bouts), while
intrabout pauses are short pauses between vocalization syllables
within bouts (22). Interbout and intrabout pauses were deter-
mined by cutoffs calculated according to histograms of pause
distributions (SI Appendix,Fig.S2B) (11, 22, 25). Previous studies
showed that a key factor contributing to the increased pause du-
rations in the vocalizations of Gnptab mut/mut animals is an in-
crease in the interbout pause durations (22). In Ser321Gly mut/
mut animals, the interbout pause durations were significantly
longer than those of wild-type littermates (P=0.00056, nof syl-
lables >15) (Fig. 1A). Animals with very low levels of vocalization
overall produced data with very large apparent interbout pause
values. To address potential bias from the inclusion of such ani-
mals, we also analyzed the USV recording data using a higher
syllables cutoff value. A significant increase in pause durations was
detected in mut/mut animals compared to wild-type littermates
using this higher syllable cutoff as well (P=0.040, nof syllables >
50) (Fig. 1B). As expected from the increase in the duration of
long pauses, the number of bouts per recording was significantly
decreased in the mut/mut animals compared to their wild-type
littermates (SI Appendix,Fig.S3Aand D). However, in contrast
to previous studies, we also found that the Ser321Gly mut/mut
mice showed a small but significant increase in the duration of
short pauses within vocalization bouts, known as intrabout pauses
[mean of
=0.149, mut/mut =0.156, P=0.010 for nof sylla-
bles >15 (Fig. 1E); mean of
=0.149, mut/mut =0.154, P=
0.040 for nof syllables >50 (Fig. 1F)]. The total number of syl-
lables produced by Ser321Gly mut/mut animals was significantly
less than that of their wild-type littermates (SI Appendix,Fig.S3B
and E). This was due to longer pause lengths between vocaliza-
tions, because vocalization durations (syllable lengths) did not
differ between mutant and wild-type groups (SI Appendix,Fig.S3
Cand F). Increased pause durations were observed in the mut/
mut animals in 4 replicate trials. Measures of increased duration
displayed a trend toward significance in trials 2 and 4, and reached
significance in trials 1 and 3 (SI Appendix, Fig. S4), supporting the
conclusion that Gnptab Ser321Gly mut/mut mice produce vocali-
zations with longer pauses.
The vocalizations of the Gnptab Ala455Ser mice displayed a
slightly different phenotype. They did not show a significant
difference when we used 15 as the minimal syllable cutoff value
(Fig. 1C), although this value did become significantly different
when we used 50 as syllables as our cutoff (P=0.0027) (Fig. 1D).
Intrabout pause duration analysis did not reach statistical sig-
nificance (Fig. 1 Gand H).
In addition to differences in vocalization timing, our previous
study showed that Gnptab mut/mut mice exhibit higher stereo-
typy in the temporal sequencing of their vocalizations, as revealed
by reduced temporal entropy in their vocalizations (22). To test
whether Ser321Gly mut/mut mice also have such vocalization
abnormalities, we analyzed both the usage and temporal sequenc-
ing of syllable types as categorized by an established classification
scheme based on presence and size of abrupt pitch jumps (Fig. 2A)
(11). As in the previous study, we found no statistically significant
differences in the syllable types and their rates of usage between
Ser321Gly mut/mut and their wild-type littermates (Fig. 2B). We
tested the temporal diversity of the sequence of vocalizations using
a first-order Markov process model, which compares the entropy
present in the temporal structure of these vocalizations. We found
significantly decreased entropy in the temporal structure in
Ser321Gly mut/mut compared to wild-type (P=0.0064) (Fig. 2C),
which indicates that the Ser321Gly mut/mut mice exhibit greater
stereotypy in temporal sequence, consistent with findings in pre-
vious studies. A higher degree of stereotypy in vocalizations can
result from an increased repetition of syllables. We found no
significant increase in the presence of doublet repetitions in the
vocalizations of Ser321Gly mut/mut. However, there may be a
trend toward an increased percentage of such doublets in the
Ser321Gly mutant animals (P=0.13, mean of
=0.591, mut/
mut =0.638) (Fig. 2D). A similar analysis was performed on the
vocalizations of the Ala455Ser mice. These analyses showed that
the mut/mut animals also displayed slightly reduced temporal
entropy, although this difference did not reach statistical signifi-
cance (SI Appendix,Fig.S5).
Taken together, the above findings indicate that mice carrying
human stuttering-associated mutations produce vocalizations
with longer pauses between syllables or bouts of syllables, and
exhibited higher stereotypy in the temporal sequence of their
vocalizations compared to wild-type littermates. These findings
are consistent with the abnormal vocalization phenotype previously
found in mutant mice carrying a different human stuttering-
associated mutation (22). Because the vocalization phenotype in
the Ser321Gly is more pronounced than that in Ala455Ser mouse,
we focused on Ser321Gly mice for subsequent studies on the
neuropathology of stuttering caused by this mutation.
Plasma Levels of Several Acid Hydrolases Are Elevated in Gnptab
Ser321Gly Mice. In humans, homozygous loss-of-function muta-
tions in GNPTAB cause mucolipidosis types II and III, in which
the lack of the mannose 6-phosphate targeting signal results in
abnormal accumulation of lysosomal enzymes in the plasma (26,
27). Mice totally deficient in GlcNAc-1-phosphotransferase ac-
tivity display several symptoms similar to those seen in humans
with mucolipidosis types II and III. For example, Gnptab
knockout mice exhibit a 7- to 14-fold increase in plasma levels
of lysosomal acid hydrolases when the ability to synthesize the
Man-6-P recognition marker is missing (27). To assess whether
missense mutations found in human stuttering give rise to a similar
phenotype in mice, the activity of 5 acid hydrolases was measured
in the plasma of the Ser321Gly mice and compared with wild-type
| Han et al.
Downloaded by guest on October 7, 2020
(Table 1). Although not as high as that in knockout mice, Gnptab
Ser321Gly animals displayed a significant 1.26- to 3.3-fold increase
in the plasma activity level of over that in wild-type mice for
β-hexosaminidase, α-mannosidase, and β-mannosidase in both
male and female groups (Table 1). There was no significant dif-
ference in the plasma activity levels of β-galactosidase and
β-glucuronidase between Ser321Gly and wild-type mice. These data
indicate that these missense mutations have functional effects on the
biological activity of the lysosomal targeting function in these mice.
Nonvocal Behaviors in Gnptab Ser321Gly Mice. To examine whether
Ser321Gly mut/mut mice exhibit any phenotypic anomalies be-
yond impaired vocalizations, we performed a battery of non-
vocalization behavioral tests on adult animals. First, we measured
spontaneous locomotion and exploratory behaviors in a 1-h open-
field test. In previous studies, it was shown that Gnptab Glu1179-
Lys homozygous mutant mice exhibited changes on 2 nonvocal
behavioral measures, spontaneous locomotion and olfactory ex-
ploration (22), although these were not significantly different from
controls when a Bonferonni correction was applied for the multi-
ple behavioral tests performed. Here, we first measured sponta-
neous locomotion and exploratory behaviors in a 1-h open-field
test. We found no significant genotypic differences in locomotor
activity in Ser321Gly and wild-type animals (Fig. 3A). Measures of
grip strength and prepulse inhibition of acoustic startle response
also failed to show any significant differences between genotypes
(Fig. 3 Band E). To measure balance and motor function, accel-
erating rotarod tests were performed over 3 successive days. Two-
way ANOVA revealed no significant genotype effect (P=0.28)
(Fig. 3C), and the KolmogorovSmirnorv test also showed no
significant difference between the 2 genotype groups (P=0.16).
We then measured nose poking and olfactory exploration of
Ser321Gly mice for both empty and odorant-containing holes
compared to wild-type littermates (22), using chambers having
both novel and familiar odorants. The duration of time spent with
home-cage materials (familiar odor) or coconut (novel odor) was
calculated and compared between wild-type and mutant mice.
Two-way ANOVA revealed that both wild-type and Ser321Gly
mice spent more time exploring the familiar odor than the novel
odor (P<0.0001), but this preference did not differ by genotype
(Fig. 3D).Thesedatasuggestthatthisstuttering-causal mutation
does not have any effect on exploratory activity guided by olfactory
We next examined social preference in our mice by measuring
their preference for social compared to nonsocial stimuli in a 3-
chamber experiment. We found that Ser321Gly and wild-type
mice both robustly preferred interacting with a novel mouse
placed in a containment cup compared to an identical cup that
Fig. 1. Increased pause durations in the vocalizations of Gnptab Ser321Gly (S321G) and Ala455Ser (A455S) compared to wild-type littermates. All graphs
indicate pause lengths in vocalizations of individual animals (1 point =1 animal). The average of individual mean pause lengths in each genotype group was
compared by ttest calculating 2-tailed Pvalue. Green represents wild-type (
), blue represents heterozygous knockin (+/mut), red represents homozygous
knockin (mut/mut). All Pvalues depicted in the figures are comparison between
and mut/mut. *P0.05; **P0.01; ***P0.001. Error bars indicate the
SEM. (A) Interbout pause duration analysis in S321G animals displaying n>15 syllables. Sample sizes (n)of
=46, nof +/mut =75, nof mut/mut =40. (B)
Interbout pause duration analysis with animals displaying total n>50 syllables. The nof
=45, nof +/mut =71, nof mut/mut =32. (C) Interbout pause
duration analysis in A455S animals displaying n>15 syllables. The nof
=31, nof +/mut =63, nof mut/mut =30. (D) Interbout pause duration analysis in A455S
animals displaying n>50 syllables. The nof
=29, nof +/mut =62, nof mut/mut =30. (E) Intrabout pause duration analysis in S321G animals displaying total n>
15 syllables. The number of animals of eachgenotypearethesameasthatinA.(F) Intrabout pause duration analysis in S321G animals displaying total n>50 syllables.
The number of animals of each genotype are the same as that in B.(G) Intrabout pause duration analysis in A455S animals displaying total n>15 syllables. The
number of animals of each genotype are the same as that in C.(H) Intrabout pause duration analysis in A455S animals displaying total n>15 syllables. The number of
animals of each genotype are the same as that in D. Degrees of freedom (DF) in Aand Eis 84, in Band Fis 75, in Cand Gis 59, in Dand His 57.
Han et al. PNAS Latest Articles
Downloaded by guest on October 7, 2020
did not contain a mouse, but there was no significant difference
of interacting period between wild-type and mut/mut mice (Fig.
3F). Together, these results suggest that Ser321Gly does not cause
global motor or cognitive deficits that could contribute to the
impaired vocalizations that we observed.
Neuropathology Studies and Reduced Astrocyte Staining Density in
Gnptab Ser321Gly Mice. Given the reproducible vocalization phe-
notype present in mice carrying different Gnptab mutations ini-
tially identified in humans who stutter, we sought to identify a
specific neuropathology in these mice. Because the vocalization
phenotype in the Ser321Gly is more pronounced than that in the
Ala455Ser mice, we focused on Ser321Gly mice for subsequent
studies of neuropathology associated with the vocalization deficit
caused by this mutation. Because a histopathologic survey of brain
and other tissue sections stained with H&E revealed no striking
differences between homozygous mutant animals and wild-type,
we investigated the brains of Ser321Gly mut/mut mice using im-
munohistochemistry with cell type-specific antibodies. These were
chosen to interrogate major brain cell types and brain areas pre-
viously suggested to be sites of the primary neurological deficit in
stuttering. Our first hypothesis was chosen based on imaging
studies of the brains of humans who stutter, which have detected
associations between stuttering and white matter deficits, in-
cluding altered connectivity (10, 28, 29). White matter contains no
neuronal cell bodies or synapses but does contain tightly packed
glial cells, including oligodendrocytes (which produce myelin),
astrocytes, and microglia. Astrocytes have been associated with
developmental white matter disorders (30, 31), and pathological
studies of Gnptab knockout mice showed demyelination (32) and
microgliosis (32, 33). Based on this, we investigated the 3 glial cell
typesastrocytes, oligodendrocytes, and microglial cellsin our
Gnptab Ser321Gly mut/mut mice. Another major hypothesis re-
garding the genesis of stuttering involves functions of the basal
ganglia (34, 35). Dopaminergic neurons form a major component
of this brain region, and dopaminergic agents, such as risperidone
and olanzapine, have been reported to improve fluency in some
subjects who stutter (3638). To investigate this, we studied DRD2
dopaminergic neurons in our mice. Yet another area of interest is
the cerebellum, which is involved in fine motor control, and there
have been reports that the cerebellum is involved in the neural
pathway for speechproduction(3941). In addition, a mouse model
of developmental verbal dyspraxia caused by a mutation of Foxp2
has been shown to have deficits of cerebellar Purkinje cells (42).
Based on these findings, we also investigated cerebellar Purkinje
cells in our mice. Brain tissues were stained with antibodies that
include anti-glial fibrillary acidic protein (Gfap) to visualize astro-
cytes, anti-myelin basic protein (MBP) to visualize oligodendro-
cytes, anti-ionized calcium binding adaptor 1 (Iba1) to identify
microglial cells, anti-calbindin (Calb1) to identify Purkinje cells of
the cerebellum, and anti-tyrosine hydroxylase (TH) to identify do-
paminergic neurons in the basal ganglia.
Astrocytes visualized by immunostaining with an anti-Gfap
antibody were studied in the corpus callosum (CC) and hippo-
campal areas, which are known to have abundant astrocyte
populations. To provide an anatomical landmark that allowed
us to match the sectional position across brains, we compared
hippocampal pyramidal layer staining in each brain section, and
chose sections with similar morphology between mutant and
wild-type (SI Appendix, Fig. S6). Coronal sections of brains from
5 mutant and wild-type pairs of P8 mice were used for staining
and quantitation. We found the anti-Gfapstained area, which
represents the amount of astrocytes, was significantly reduced in
Fig. 2. Reduced entropy in the sequence of vocalizations of Gnptab
Ser321Gly mice compared to wild-type littermates. (A) Sonograms of dif-
ferent syllable types. Syllable types were defined as previously described
(20). (B) Percentage of times that 1 syllable type was followed by the same
syllable type. Each color represents 1 syllable type as defined in A.(C)Di-
versity of vocalization sequences as quantified by the entropy of the cor-
responding first-order Markov process as previously described (20). (D)
Proportion of repeated syllables (%). (BD) Analyses of animals with syllable
n>15. The
and mut/mut were compared by ttest calculating 2-tailed P
value in Cand D. Error bars indicate the SEM. **P0.01. Samples sizes of
each genotype are the same as that in Fig. 1 Aand C.DFinCand Dis 84.
Mean entropy of
in Cis 1.29 and that of mut/mut in Cis 1.09.
Table 1. Lysosomal enzyme activity in the plasma of Gnptab Ser321Gly mice
Animal group β-Galactosidase β-Glucuronidase β-Hexosaminidase β-Mannosidase α-Mannosidase
Wild-type (n=9) 24.95 ±5 9.20 ±1.8 206.73 ±20 52.26 ±3.7 117.33 ±20.9
Ser321Gly (n=8) 23.00 ±6 7.72 ±1.4 260.88 ±35 112.39 ±22.9 279.34 ±77.1
Pvalue 0.46 0.09 0.001 <0.0001 <0.0001
Wild-type (n=7) 24.49 ±3.9 8.70 ±1.1 237.32 ±17.3 36.12 ±1.7 83.51 ±10.8
Ser321Gly (n=7) 28.00 ±5.5 10.20 ±2.0 390.81 ±105.8 90.78 ±11.1 272.3 ±22.7
Pvalue 0.19 0.12 0.0025 <0.0001 <0.0001
Wild-type 24.75 ±4.5 8.98 ±1.6 220.11 ±24.2 45.20 ±8.8 102.53 ±24.1
Ser321Gly 25.34 ±5.9 8.87 ±2.1 321.51 ±99.5 102.30 ±21.0 276.05 ±56.6
0.38 0.44 <0.001 <0.0001 <0.0001
Activity is expressed as nanomole 4-methyumbelliferone cleaved per hour per milliliter of plasma.
| Han et al.
Downloaded by guest on October 7, 2020
the CC of the Ser321Gly mut/mut mice compared to that of wild-
type littermates in all 5 pairs tested (SI Appendix, Fig. S7A; see
also Fig. 5A). Quantification of the stained images shows that the
difference in anti-Gfapstained area between mutant and wild-
type group is significant in both the left and right CC (fold-
change in left: 26.3%; P=0.013, fold-change in right: 19.5%, P=
0.017) (Fig. 4A). The anti-Gfapstained area in the hippocampal
subfield CA1 of Ser321Gly mut/mut was also lower than in wild-
types, but this difference was smaller than that in the CC. (SI
Appendix, Fig. S7B)
Because homozygous GNPTAB deficiency in humans and
mice leads to progressive degenerative effects on the nervous
system (32, 33), we performed the same anti-Gfap immunos-
taining in 3 pairs of older (16 mo of age) Ser321Gly mut/mut
animals and their wild-type littermates, to see if there were any
age-associated degenerative effects on the amount of astrocyte
staining. Results from these 3 brain pairs showed a lower amount
of astrocyte staining in CC of Ser321Gly mut/mut animals com-
pared to wild-type, and this difference was highly significant (Fig.
4Band SI Appendix,Fig.S8). This difference was greater than that
in P8 mice, suggestive of an astrocyte deficit that worsens with age
in these mice.
No Differences in Staining of Other Brain Cells in Gnptab Ser321Gly
Mice. Next, we measured the density of microglial cells in the CC
and hippocampal areas of P8 mice using immunostaining with
anti-Iba1 antibody. Anti-Iba1 staining area in the CC was similar
to that in the hippocampus (SI Appendix, Fig. S9) and there was
no difference in the cell area stained between Ser321Gly mut/mut
and wild-type littermates (Fig. 5Aand SI Appendix,Fig.S9). We
also studied these tissues in adult animals (16 mo of age) and
found no difference in the amount of anti-Iba1 staining area be-
tween the mutant and wild-type groups (SI Appendix,Fig.S10A).
Anti-MBP antibody was used to stain myelin-containing oli-
godendrocytes in the CC area. We found no significant differ-
ence in the area stained with anti-MBP between P8 homozygous
mutant animals and their wild-type littermates (Fig. 5B). In adult
mice, the anti-MBP staining was too saturated (SI Appendix, Fig.
S10B) to quantify. To address this, we also used Luxol fast blue
staining to visualize myelin in adult mouse brains and found no
significant difference in these areas between mutant and wild-
type mice (SI Appendix, Fig. S10C).
Fig. 4. Reduced astrocyte staining in the CC of Gnptab Ser321Gly mouse
brains. Perfusion-fixed coronal cryosections (10-μm thickness) were used for
immunostaining using anti-Gfap staining for astrocytes in the CC area.
Quantification of the stained area was done using ImageJ software and
paired ttests were used to test statistical significance of staining differences
between genotype groups by calculating 2-tailed Pvalues. (A) Immunos-
taining and quantitation of the stained area in P8 mice. Sample sizes (n)
=10, nof mut/mut =10,DFis9forbothhemispheres.(B) Immunostaining
and quantitation of the stained area in 16-mo-old mice. The nof
=8, n
of mut/mut =8, DF is 7 for left hemispheres. The nof
=9, nof mut/
mut =8, DF is 8 for right hemispheres. (Scale bars, 100 μm.) Error bars
indicate the SEM. *P0.05; ***P0.001; ****P0.0001.
Fig. 3. Nonvocal behavioral tests of Gnptab Ser321Gly mice. Sixteen 7-mo-
old Ser321Gly mut/mut mice and equal numbers of sex and age-matched
mice were used for 5 different nonvocal behavior tests, and 26 mut/mut and
equal number of age-matched
mice were used for rotarod test. Black
squares or bars indicate wild-type and gray squares or bars indicate Gnptab
Ser321Gly mut/mut homozygotes. (A) One-hour open-field locomotive ac-
tivity. (B) Grip strength. (C) Rotarod test with accelerating speed. (D)Ol-
factory discrimination. (E) Prepulse inhibition. (F) Social interaction. Error
bars indicate the 95% confidence interval except for 1-h open-field loco-
motive activity test which has SD as error bars. DF is 50 for the rotarod test
and 30 for all of the other tests.
Han et al. PNAS Latest Articles
Downloaded by guest on October 7, 2020
Because stuttering represents a specialized motor control prob-
lem, we also examined cerebellar structures. We compared num-
bers of cerebellar Purkinje cell bodies in mutant and wild-type
animals using immunostaining with anti-Calb1 antibody. A
granular layer in the same lobule (the junction between lobule
IV-V and the simple lobule, of the
cerebellum was chosen for comparison between mutant and wild-
type. The number of anti-Calb1-stained cell bodies was counted
and compared between mutant and wild-type pairs. The num-
ber of Purkinje cell bodies showed no difference between the 2
genotype groups in both P8 (Fig. 5C) and adult mice (SI Ap-
pendix, Fig. S10D).
Finally, dopaminergic neurons in striatum were stained with
anti-TH antibody in P8 mice. Because the staining tended to be
saturated across a network of nerve cells in this tissue, it was
difficult to identify a difference at the level of light microscopy
between the 2 genotype groups. However, quantification analysis
showed no significant difference of staining area between the 2
genotypic groups in both P8 (Fig. 5D) and adult mice. (SI Ap-
pendix, Fig. S10E).
Detection of a White Matter Deficit in Gnptab Ser321Gly Mice. Be-
cause diffusion tensor imaging (DTI) has been shown to be
sensitive to subtle differences in the white matter tissue envi-
ronment, and because DTI scalars are known to change during
postnatal white matter development and maturation in humans
and rodents (43), we performed high-resolution, high-quality
14.1 Tesla DTI mapping in whole-brain fixed tissue specimens
from P8 Ser321Gly mut/mut mice and their wild-type littermates.
Details of DTI scanning and analysis procedures were described
in SI Appendix,SI Materials and Methods. There were no gross
morphologic differences between Ser21Gly mut/mut and wild-
types by brain qualitative inspection (SI Appendix, Fig. S11),
and most of DTI valuessuch as fractional anisotropy, trace,
axial diffusivity, and radial diffusivity after registration were not
significantly different between mut/mut and
littermates (Fig.
6). However, the template-based region-of-interest (ROI) anal-
ysis of LogJ showed significantly reduced local volume of the
genu of the CC (mean ±SEM in
=0.017 ±0.011, mut/
mut =0.16 ±0.012, P=0.025) and no significant difference of
the anterior commissure (P=0.26), the splenium (P=0.67), and
the hippocampus (P=0.97) of mut/mut brains compared to those
of their
littermates (Fig. 6).
Abnormal Vocalization in Astrocyte-Specific Gnptab Knockout Mice.
Our immunostaining data suggested that mice that carry human
stuttering mutations and have abnormal vocalizations also have
alterations specifically in their brain astrocytes. To indepen-
dently test this hypothesis, we knocked out the Gnptab gene in
specific brain cell types using mice that express Cre recombinase
under the control of different brain cell type-specific promoters.
We constructed an astrocyte-specific knockout of Gnptab as well as
oligodendrocyte-specific, Purkinje cell-specific, and dopaminergic
neuron-specific knockouts of Gnptab, and measured the vocaliza-
tions in P8 animals carrying these cell type-limited mutations.
We chose brain-specific Cre-driver mouse lines from the
GENSAT Project at The Rockefeller University as follows:
Gfap-Cre for astrocyte-specific expression of Cre, Adroa2a-Cre
for dopaminergic Drd2 receptor-specific neurons in striatum,
Pcp2-Cre for expression of Cre in cerebellar Purkinje cells, and
Plp1-Cre for Cre expression in oligodendrocytes. Expression of
the Cre recombinase in each mouse line was previously confirmed
using an enhanced green fluorescent protein reporter in P7 mice
and reported in the GENSAT database (
cre.jsp)(SI Appendix,Fig.S12).
We constructed conditional knockout mice carrying the loxP
sequence in the 5and 3flanking regions of exon 2 of Gnptab.
The Gnptab with floxed exon 2 was designed to be located on
Fig. 5. No significant effect of the Gnptab Ser321Gly mutation on immunoreactivities in nonastrocyte cell types. (A) Immunostaining of microglial cells in the
CC area with anti-Iba1 and quantitation of stained area. Sample sizes (n)of
=5, nof mut/mut =5, DF is 4. (B) Immunostaining of oligodendrocytes in CC
area with anti-MBP and quantitation of stained area. The nof
=6, nof mut/mut =6, DF is 5. (C) Immunostaining of cerebellar Purkinje cells with anti-
Calb1 and quantitation of stained area. The nof
=6, nof mut/mut =6, DF is 5. (D) Immunostaining of dopaminergic neurons with anti-TH and quan-
titation of stained area. The nof
=8, nof mut/mut =8, DF is 7. (Scale bars, 100 μm.) Age of all mice is P8. All data were obtained from left hemispheres of
coronal sections. Error bars indicate the SEM.
| Han et al.
Downloaded by guest on October 7, 2020
one chromosome, while the other chromosome had either the wild-
type or fully deleted exon 2 in order to maximize the recombinase
efficiency for deletion of the exon in the specific cells (SI Appendix,
Fig. S13). Pup isolation calls at day P8 were recorded and analyzed
using the same methods used for those of Ser321Gly and Ala455Ser
knockin mice.
Gfap-specific Gnptab knockout mice showed significantly in-
creased interbout pause durations in the conditional heterozy-
gote knockouts (cHET) (P=0.05), and a trend toward increased
interbout durations in conditional homozygote knockouts
(cHOM), although this difference did not reach statistical sig-
nificance (P=0.09, mean of wild-type =2.33, cHOM =3.04)
(Fig. 7A). To confirm the astrocyte-specific knockout of Gnptab
in mutant mice, we examined the colocalization of anti-Gnptab
staining and anti-Gfap staining in P8 mice of cHOM and wild-
type. In wild-type mice, anti-Gnptab staining labeled 36.6 ±3.1%
of the Gfap-staining area, but the anti-Gnptab staining labeled
15.3 ±2.4% of the Gfap-staining area in cHOM mice (Fig. 7E).
This difference between wild-type and cHOM was highly signifi-
cant (P=0.0001) (Fig. 7E), and supports the conclusion that
Gnptab was selectively knocked out in the astrocytes of these mice.
We found no significant differences in interbout pause dura-
tions between oligodendrocyte, Drd2 neuron, and Purkinje cell-
specific Gnptab knockout mice and their wild-type littermates
(Fig. 7 BD). Thus, 2 different experimental approaches, im-
munohistochemistry and brain cell type-specific expression of
Gnptab mutations, indicate that a deficit in astrocytes, and not in
other brain cell types previously proposed to be involved in the
etiology of stuttering, underlies the altered vocalizations in mice
carrying human stuttering mutations.
Stuttering is characterized by interruptions in the flow of speech.
The temporal characteristics of pauses in USVs of mutant mice
are similar to those identified in the speech of humans who
stutter and carry mutations in the GNPTAB gene (22). We show
here that, in addition to the single mutation studied previously, 2
other Gnptab mutations initially identified in humans who stutter
(Ser321Gly and Ala455Ser) also result in abnormal pauses in the
USV of mice who carry them. These mice also display reduced
temporal entropy in their vocalizations, consistent with increased
stereotypy in these vocalizations. Thus, mice carrying human
stuttering mutations in Gnptab present an animal model that
recreates 2 core phenotypic features of human stuttering. Mouse
models of other inherited speech disorders have been created,
for example using mutations in FOXP2, and these mutations cause
an absence of vocalizations in pups (42, 44). More recent work has
shown that a Foxp2 mutation can produce a vocalization syntax
deficit in adult male mice, with deficits in the production of more
complex USV sequences in social contexts, and that this is ac-
companied by a shift in the position of the laryngeal motor cortex
layer-5 neurons in these animals (19). Humans and song-learning
birds share characteristic features that include a forebrain system
for vocal modification and auditory feedback (45). Because mice
have been shown to possess limited vocal modification ability and
some neuroanatomical features found in humans and song-
learning birds (18), these findings and the present study show
that mice can be used as model animals to investigate human
speech disorders and also suggest a deep evolutionary conserva-
tion of some of the neural mechanisms involved in vocalization.
In humans, homozygous loss-of-function mutations in GNPTAB
cause mucolipidosis types II and III, which are lysosomal storage
disorders characterized by an abnormal presence of lysosomal
hydrolases in the circulating plasma (4648). In our Gnptab
Ser321Gly mice, we found abnormal plasma levels in 3 of the ly-
sosomal hydrolases tested. In contrast to mucolipidosis patients,
humans who stutter and carry mutations in this gene typically
have missense mutations and do not display any symptoms of
mucolipidosis (3, 4). Although the mice carrying the missense mu-
tation displayed moderate elevations of these lysosomal hydrolases,
there were no signs of mucolipidosis in a wide range of tissues, and
other than their USVs, their behavior was normal in a wide range of
behavioral assays.
We have investigated the pathology that underlies the vocal-
ization deficit in these mice at molecular, cellular, and anatomic
scale. While an animal-wide pathology evaluation failed to show
a difference in the Gnptab knockin animals by traditional H&E
staining, a more detailed immunohistochemical study of the brains
of these animals revealed one significant difference between them
and their wild-type littermates, which was a deficit in staining with
an anti-Gfap antibody, a classic marker of astrocytes. This deficit
was especially prominent in the CC. Notably, we found no dif-
ference in staining with anti-Calb1, which is specific for Purkinje
cells that are a major component the cerebellum. Deficits in this
brain region, which plays a major role in the control of motor
function, are prominent in mice carrying FOXP2 mutations (42).
The normal staining of Purkinje cells in our mice further rein-
forces the view that stuttering is both genetically (49) and patho-
logically distinct from developmental verbal dyspraxia, such as that
associated with FOXP2 mutations.
The involvement of an abnormality in astrocytes in the mouse
vocalization deficit was supported by our studies of neural Cre-
driver lines. We generated 4 such conditional knockout lines that
expressed this Gnptab mutation solely in astrocytes, oligodendro-
cytes, Purkinje cells, or dopaminergic neurons, respectively. Of
these, only the astrocyte-specific Gnptab knockout displayed a
vocalization phenotype that was significantly different from their
wild-type littermates. This phenotype consisted of more pauses and
abnormally long pauses in their vocalizations, similar to the vo-
calization deficit in the whole-animal Gnptab knockin mutations.
Astrocytes have a wide range of critical roles in brain ho-
meostasis (50), which include nutrient support for neurons (51),
uptake and modulation of synaptic transmitter such as glutamate
(52, 53), regulation of potassium concentration of extracellular
space (54), and nervous system repair (55). While astrocyte
deficits have not previously been suggested to be associated with
stuttering, recent results have shown that these cells play an active
role in the pathogenesis of neurological disorders (30, 31, 56). One
of these studies reported that astrocytes are involved in patho-
logical activation of lysosomal disorders by a noncell-autonomous
functional pathway (30), which would be consistent with astrocytes
Fig. 6. Detection of a subtle white matter abnormality in Gnptab Ser321Gly
mice by DTI. Quantitative analysis of template-based ROI values for frac-
tional anisotropy (FA), Trace, and LogJ in the anterior commissure (AC), genu,
splenium, and hippocampus are shown. Lines connect littermate pairs with the
same sex and different genotype in the study. Three
and mut/mut pa irs
were marked as square, triangle, and circle. Paired t-tests were used to test
significance by calculating 2-tailed Pvalues. LogJ at the genu was significantly
different between
and mut/mut. *P0.05. DF of each analysis is 2.
Han et al. PNAS Latest Articles
Downloaded by guest on October 7, 2020
being involved in pathogenesis of human stuttering caused by a
deficit in lysosomal targeting functions. Previous studies have
shown that levels of astrocytes and microglial cells are increased in
adult mice carrying a knockout mutation of Gnptab,whichis
model of mucolipidosis type II (32, 33). These studies demon-
strated that astrogliosis and inflammation causing neurodegeration
are activated by this complete knockout of Gnptab, which is as-
sociated with systemic pathologies. However, our data showed that
astrocyte levels of 16-mo-old mutant mice carrying Gnptab
Ser321Gly are still lower than those in wild-type littermates. This
result suggests that partial deficit of lysosomal targeting function
caused by a missense mutation in Gnptab is not associated with
increased astrogliosis and inflammation, but rather is associated
with a loss of other physiological functions of astrocytes.
The most prominent location of astrocyte pathology we observed
by immunohistochemistry was in the CC. The CC connects the
2 hemispheres of the brain, and abnormalities in interhemispheric
functions has been previously suggested to be associated with
stuttering (57). DTI analysis also suggested a potential reduction in
local volume of the CC in Gnptab Ser321Gly mice. A location for
the site of the functional deficit in the CC in our mutant mice is
consistent with recent results from brain structural imaging studies
in children who stutter, a recent study of which showed abnor-
malities of connectivity in the CC. These were particularly notable
in children with persistent rather than recovered stuttering (29).
The human subjects with documented GNPTAB mutations to date
all have persistent stuttering (3, 4). In addition, studies of our an-
imal models show that astrocyte pathology persists into adulthood
in these animals. Thus, human genetic findings (2), animal model
studies, and independent human brain imaging studies (3, 4) now
all support a role for a cellular deficit in the CC in the genesis of
persistent developmental stuttering.
The present study suggests astrocytes as a previously unsus-
pected site of the pathophysiology underlying human stuttering.
Because Gnptab is universally expressed in brain cells and tissues
(49, 58, 59), a current question is why astrocytes present an in-
creased susceptibility to pathology from partial loss of function
mutations in Gnptab compared to other brain cell types. We
note that our data do not completely rule out the possibility
that other brain cells are involved in pathogenesis of stutter-
ing associated with mutations in Gnptab (22). Also unknown
are the precise mechanisms by which this Gnptab functional
Fig. 7. Pause duration analysis in the vocalizations of brain cell type-specific Gnptab knockout mice compared to wild-type littermates. The average of
individual mean pause lengths in each genotype group was compared by ttest calculating 2-tailed Pvalue. *P0.05; ****P0.0001. Error bars indicate the
SEM. (A) Interbout pause duration analysis In Gfap Cre-driver (astrocyte-specific) Gnptab knockout mice. Sample sizes (n) of wild-type animals =29, nof cHET
animals =13, and nof cHOM animals =15. (B) Interbout pause duration analysis in Plp1 Cre-driver (oligodendrocyte-specific) Gnptab knockout mice. The nof
wild-type animals =10, nof cHET animals =8, nof cHOM animals =10. (C) Interbout pause duration analysis in Adora2a Cre-driver (Drd2 neuron-specific)
Gnptab knockout mice. The nof WT animals =12, nof cHET animals =8, nof cHOM animals =11. (D) Interbout pause duration analysis in Pcp2 Cre-driver
(Purkinje cell-specific) Gnptab knockout mice. The nof WT animals =8, nof cHET animals =9, nof cHOM animals =8. Number of syllables cutoff for ADis 15.
(E) Immunostaining of CC area with anti-Gfap (red) and anti-Gnptab (green) in cHOM (Bottom) and wild-type littermates (Top) and quantification of stained
area. White color indicates areas of colocalization, which was analyzed using the Colocalizationplugin (
in ImageJ software with default threshold. Quantification of immunostaining area in which ratio of colocalization area versus Gfap staining area was calculated.
(Scale bars, 100 μm.) Sample size for quantification of immunostaining is depicted on top of each bar graph.
| Han et al.
Downloaded by guest on October 7, 2020
deficit leads to our observed difference in astrocyte cellular
density. However, our evidence for the involvement of this cell
type and specific brain structure present new avenues for the
understanding of cause of stuttering at anatomic, cellular, and
molecular scales within the brain.
Materials and Methods
Animal Subjects. Mice were housed with 1 to 5 animals per cage in standard
plastic home cages on a 12-h light/dark cycle, and all vocal or nonvocal be-
havioral experiments were performed during the light phase. All mice used in
this study have C57BL/6J strain background. All experiments and mainte-
nance of mice were conducted in accordance with the National Institutes of
Health Guidelines for the Care and Use of Laboratory Animals, and were
carried out under the National Institute on Deafness and other Communi-
cation Disorders Animal Care and Use Committee protocol #1318-16. One to
5 mice were housed per cage. In case of 5 males in a cage, they were sep-
arated into multiple cages at sexual maturity (6 to 8 wk) to prevent fighting.
Mouse Vocalization Recording and USV Analysis. USVs of mice were recorded
on P8 using a recording system from Avisoft Bioacoustics (https://www.avisoft.
com). A heterozygote ×heterozygote breeding strategy was used to en-
sure the availability of wild-type littermate controls with matched pre- and
postnatal environment. Foot tattoos were applied for temporary identification
of pups 1 to 3 d before recording, prior to genotyping the animals. Animal
vocalizations were recorded and vocalization data were analyzed blind to ge-
notype. Before recording of pups, the dam mice were removed from their pups
and the cage was kept on a warm heating pad during experiments. After an
initial 10-min equilibrium period on the heating pad, the pup was transferred to
the recording chamber and the body temperature of each pup was measured
using an infrared thermometer. Vocalization recordings of each pup were done
in a plastic 12.4-cm ×12.4-cm ×13.8-cm chamber for 5 min. After identification
using the foot tattoo and measuring body weight, pups were returned to the
dam. The pups were genotyped after weaning on P21. The USV data were
analyzed using Matlab (Mathworks, including
the signal-processing toolbox, image-processing toolbox, statistics and machine
learning toolbox. For quantitative analysis, Matlab codes originally developed
by T. Holy (11) and subsequently modified by T. Barnes (22) we re used. Mean
pause lengths between syllables (intrabout) or groups of syllables (interbout) in
vocalizations of each genotype group were compared by ttest and 2-tailed P
value was used in each analysis. All experiments were done in 4 different stages,
and both data of each stage and combined analysis data are shown.
Nonvocalization Related Behavior. Thirty-two mice, 16 Gnptab Ser321Gly mut/
mut and 16 wild-type littermates, were tested at 7 mo of age on a battery of
nonvocalization-related behaviors to assess spontaneous locomotion, fore-
limb strength, motor coordination, novelty seeking and aversion, social
recognition memory, and paired-pulse inhibition. For the rotarod test, 10
Gnptab Ser321Gly mut/mut and 10 wild-type littermates were additionally
tested. Experimental procedures for behavior tests are similar to previously
described methods (22). Because mouse behaviors have been shown to be
comparable in both light and dark phase (60), vocalization and behavioral
tests were carried out exclusively in light phase to reliably standardize light
exposure across all tests. The results are shown in Fig. 4 in the order that the
tests were performed. Details of test procedures are described in SI Ap-
pendix,SI Materials and Methods.
Assay of Plasma Lysosomal Hydrolase Levels. Lysosomal hydrolase activities in
plasma samples were determined by fluorometric enzyme assays as described
previously (27). In brief, plasma samples were incubated with 5 mM 4-
methylumbelliferyl-coupled specific substrates in a 50 mM citrate buffer
containing 0.5% Triton X-100, pH 4.5, at 37 °C. Reactions were stopped by
addition of 0.1 M glycine-NaOH solution, pH 10.3, and the fluorescence read
at 495 nm. The plasma activities are expressed as nanomoles of hydrolyzed
methylumbelliferone per hour per milliliter of plasma.
Antibodies and Immunostaining. Mice of the indicated ages were anesthetized
and perfused through the heart with PBS and 4% paraformaldehyde (Elec-
tron Microscopy Sciences) and postfixed by immersion in 4% paraformaldehyde
overnight at 4 °C. The brains were cryoprotected with PBS solution containing
sucrose for 2 d (4 h in 10%, overnight in 20%, 24 h in 30% sucrose). Brain tissue
mounts were prepared for cryosectioning using OCT media with isopentane
solution chilled on dry ice. Ten-millimeter tissue sections were obtained using
Cryostat (Leica CM3050S), mounted on positively charged glass slides (Superfrost
Plus Gold; Thermo Fisher Scientific), then dried at room temperature for 20 min.
The tissue sections were stored at 80 °C until usage. Reagents for
immunostaining were prepared according to methods described in IHC-
World ( The following
antibodies and dilutions were used for the immunostaining: rabbit anti-Gfap
antibody (ab7260, 1:5,000; Abcam), rabbit anti-MBP antibody (ab40390,
1:1,000; Abcam), rabbit anti-Iba1 antibody (NBP2-19019, 1:100; Novus Bio-
logicals), rabbit anti-TH antibody (ab112, 1:500; Abcam), rabbit anti-Calb1
antibody (ab49899, 1:100; Abcam), rabbit anti-Gfap antibody (BS-13476R,
1:200; Bioss Antibodies), chicken anti-Gfap antibody (ab4674, 1:4,000;
Abcam), goat anti-Rabbit IgG H&L (Alexa Fluor 488) preadsorbed (ab150081,1:200;
Abcam), and goat anti-chicken IgY H&L (Alexa Fluor 647) preadsorbed
(ab150175,1:200; Abcam). Blocking agent treatment was performed for
30 min, followed by 16-h incubation with the primary antibodies. The anti-
bodies were washed with PBS containing 0.05% Tween20 3 times in glass
coplin jar (5 min, 15 min, 5 min). Slides were then incubated with the sec-
ondary antibody for 1 h and then washed with PBS 3 times (5 min, 15 min,
5 min). Vectashield HardSet Antifade mounting medium containing Dapi
(H-1500, Vector Laboratory) was used for final mounting after staining. An
LSM780 microscope (Zeiss) was used for the fluorescence microscopy. Quan-
tification of images was done with ImageJ ( Because
these samples represented both technical and biological replicates using a
littermate of the same sex, a paired ttest was used to test statistical signifi-
cance in difference of quantified data between genotype groups by calcu-
lating 2-tailed Pvalues.
Breeding Strategy for Brain-Cell Specific Gene Knockout Mice. To generate
conditional knockout of Gnptab in specific brain cell types, embryonic stem
cells carrying the loxP sequence fragment (ATAACTTCGTATAGCATA-
CATTATACGAAGTTAT) in the flanking regions of exon 2 of Gnptab was
purchased from EUCOMM [European Conditional Mouse Mutagenesis Pro-
(61) and germline transmission was achieved in collaboration with
TAMC, as described above. B6.FVB background B6 (C3)-Tg(Pgk1-FLPo)10Sykr/J
mice were mated with mice carrying the targeted gene cassette to remove
the neomycin cassette. Transgenic mice expressing Cre recombinase under
the control of different promoters (so called Cre-driver lines) were purchased
from Jackson Laboratory [B6.Cg-Tg(Gfap-cre)77.6Mvs/2J, B6.129-Tg(Pcp2-
Cre)2Mpin/J, B6.Cg-Tg(Plp1-cre/ERT)3Pop/J] or MMRRC [B6.FVB(Cg)-
Tg(Adora2a-cre)KG139Gsat/Mmucd]. The GENESAT database (http://www. was used to check the targeted gene expression
induced by each of the Cre-drivers (SI Appendix,Fig.S10). Heterozygote
Gnptab whole knockout mice carrying Cre gene cassette were mated with
homozygotes carrying a floxed exon 2 of Gnptab to get cell type-specific
knockout on a single chromosome, which results in 4 different genotypes (SI
Appendix,Fig.S13). USVs were tested with P8 pups using methods described
above, and genotyping and USV analysis was done after weaning.
ACKNOWLEDGMENTS. We thank Dr. James McGehee and Patrick Diers for
expert veterinary assistance; Dr. Matthew Starost for mouse pathology
evaluations; Dr. Stuart Kornfeld for assistance with the plasma enzyme
assays; Dr. Yogita Chudasama and Kevin David Cravedi for assistance with
nonvocal animal behavior tests; the NIH mouse imaging facility, especially
Dr. Jeeva P Munasinghe, for diffusion tensor imaging experiments; and Thomas
Friedman and Doris Wu for valuable comments on the manuscript. This work
was performed under National Institute on Deafness and other Communi-
cation DisordersAnimal Study Protocol 1318-16. This work was supported by
the Intramural Research Program of the NIH, National Institute on Deafness
and other Communication Disorders under intramural Grant Z1A-000046-18
(to D.D.), by the National Institute of Mental Health Rodent Behavioral Core
and by the National Heart, Lung, and Blood Institute Animal MRI Core.
1. O. Bloodstein, N. B. Ratner, A Handbook on Stuttering (Thomson/Delmar Learning,
Clifton Park, NY, ed. 6, 2008).
2. C. Frigerio-Domingues, D. Drayna, Genetic contributions to stuttering: The current
evidence. Mol. Genet. Genomic Med. 5,95102 (2017).
3. C. Kang et al., Mutations in the lysosomal enzyme-targeting pathway and persistent
stuttering. N. Engl. J. Med. 362, 677685 (2010).
4. M. H. Raza et al., Association between rare variants in AP4E1, a component of in-
tracellular trafficking, and persistent stuttering. Am.J.Hum.Genet.97, 715725 ( 2015).
5. M. H. Raza et al., Mucolipidosis types II and III and non-syndromic stuttering are asso-
ciated with different variants in the same genes. Eur.J. Hum. Genet. 24,529534 (2016).
6. J. Neefjes, R. van der Kant, Stuck in traffic: An emerging theme in diseases of the
nervous system. Trends Neurosci. 37,6676 (2014).
Han et al. PNAS Latest Articles
Downloaded by guest on October 7, 2020
7. R. Salmelin et al., Functional organization of the auditory cortex is different in stut-
terers and fluent speakers. Neuroreport 9, 22252229 (1998).
8. P. T. Fox et al., A PET study of the neural systems of stuttering. Nature 382, 158161
9. A. L. Foundas, A. M. Bollich, D. M. Corey, M. Hurley, K. M. Heilman, Anomalous
anatomy of speech-language areas in adults with persistent developmental stutter-
ing. Neurology 57, 207215 (2001).
10. S. E. Chang et al., Anomalous network architecture of the resting brain in children
who stutter. J. Fluency Disord. 55,4667 (2018).
11. T. E. Holy, Z. Guo, Ultrasonic songs of male mice. PLoS Biol. 3, e386 (2005).
12. J. M. Grimsley, J. J. Monaghan, J. J. Wenstrup, Development of social vocalizations in
mice. PLoS One 6, e17460 (2011).
13. M. L. Scattoni, J. Crawley, L. Ricceri, Ultrasonic vocalizations: A tool for behavioural
phenotyping of mouse models of neurodevelopmental disorders. Neurosci. Biobehav.
Rev. 33, 508515 (2009).
14. F. Hoffmann, K. Musolf, D. J. Penn, Spectrographic analyses reveal signals of in-
dividuality and kinship in the ultrasonic courtship vocalizations of wild house mice.
Physiol. Behav. 105, 766771 (2012).
15. J. B. Panksepp et al., Affiliative behavior, ultrasonic communication and social reward
are influenced by genetic variation in adolescent mice. PLoS One 2, e351 (2007).
16. T. Kikusui et al., Cross fostering experiments suggest that mice songs are innate. PLoS
One 6, e17721 (2011).
17. H. Choi, S. Park, D. Kim, Two genetic loci control syllable sequences of ultrasonic
courtship vocalizations in inbred mice. BMC Neurosci. 12, 104 (2011).
18. G. Arriaga, E. P. Zhou, E. D. Jarvis, Of mice, birds, and men: The mouse ultrasonic song
system has some features similar to humans and song-learning birds. PLoS One 7,
e46610 (2012).
19. J. Chabout et al., A Foxp2 mutation implicated in human speech deficits alters se-
quencing of ultrasonic vocalizations in adult male mice. Front. Behav. Neurosci. 10,
197 (2016).
20. G. A. Castellucci, M. J. McGinley, D. A. McCormick, Knockout of Foxp2 disrupts vocal
development in mice. Sci. Rep. 6, 23305 (2016). Erratum in: Sci. Rep. 7, 39722 (2017).
21. A. Fedyna, D. Drayna, C. Kang, Characterization of a mutation commonly associated
with persistent stuttering: Evidence for a founder mutation. J. Hum. Genet. 56,8082
22. T. D. Barnes et al., A mutation associated with stuttering alters mouse pup ultrasonic
vocalizations. Curr. Biol. 26,110 (2016).
23. J. A. Harris et al., Anatomical characterization of Cre driver mice for neural circuit
mapping and manipulation. Front. Neural Circuits 8, 76 (2014).
24. S. Gong et al., Targeting Cre recombinase to specific neuron populations with bac-
terial artificial chromosome constructs. J. Neurosci. 27, 98179823 (2007).
25. J. Chabout, A. Sarkar, D. B. Dunson, E. D. Jarvis, Male mice song syntax depends on
social contexts and influences female preferences. Front. Behav. Neurosci. 9,76
26. M. Boonen, P. Vogel, K. A. Platt, N. Dahms, S. Kornfeld, Mice lacking mannose 6-
phosphate uncovering enzyme activity have a milder phenotype than mice deficient
for N-acetylglucosamine-1-phosphotransferase activity. Mol. Biol. Cell 20, 43814389
27. C. M. Gelfman et al., Mice lacking alpha/beta subunits of GlcNAc-1-phosphotransferase
exhibit growth retardation, retinal degeneration, and secretory cell lesions. Invest.
Ophthalmol. Vis. Sci. 48, 52215228 (2007).
28. S. E. Chang, D. C. Zhu, A. L. Choo, M. Angstadt, White matter neuroanatomical dif-
ferences in young children who stutter. Brain 138, 694711 (2015).
29. H. M. Chow, S. E. Chang, White matter developmental trajectories associated with
persistence and recovery of childhood stuttering. Hum. Brain Mapp. 38, 33453359
30. C. Di Malta, J. D. Fryer, C. Settembre, A. Ballabio, Astrocyte dysfunction triggers
neurodegeneration in a lysosomal storage disorder. Proc. Natl. Acad. Sci. U.S.A. 109,
E2334E2342 (2012).
31. E. Sen, S. W. Levison, Astrocytes and developmental white matter disorders. Ment.
Retard. Dev. Disabil. Res. Rev. 12,97104 (2006).
32. K. Kollmann et al., Lysosomal dysfunction causes neurodegeneration in mucolipidosis
II knock-inmice. Brain 135, 26612675 (2012).
33. R. A. Idol et al., Neurologic abnormalities in mouse models of the lysosomal storage
disorders mucolipidosis II and mucolipidosis III γ.PLoS One 9, e109768 (2014).
34. P. A. Alm, Stuttering and the basal ganglia circuits: A critical review of possible re-
lations. J. Commun. Disord. 37, 325369 (2004).
35. J. Lan et al., Association between dopaminergic genes (SLC6A3 and DRD2) and
stuttering among Han Chinese. J. Hum. Genet. 54, 457460 (2009).
36. G. A. Maguire, G. D. Riley, D. L. Franklin, L. A. Gottschalk, Risperidone for the
treatment of stuttering. J. Clin. Psychopharmacol. 20, 479482 (2000).
37. G. A. Maguire et al., Olanzapine in the treatment of developmental stuttering: A
double-blind, placebo-controlled trial. Ann. Clin. Psychiatry 16,6367 (2004).
38. A. Boyd, K. Dworzynski, P. Howell, Pharmacological agents for developmental stut-
tering in children and adolescents: A systematic review. J. Clin. Psychopharmacol. 31,
740744 (2011).
39. H. Ackermann, The contribution of the cerebellum to speech and language. Brain
Lang. 127, 315316 (2013).
40. D. E. Callan, M. Kawato, L. Parsons, R. Turner, Speech and song: The role of the
cerebellum. Cerebellum 6, 321327 (2007).
41. J. R. Booth, L. Wood, D. Lu, J. C. Houk, T. Bitan, The role of the basal ganglia and
cerebellum in language processing. Brain Res. 1133, 136144 (2007).
42. E. Fujita et al., Ultrasonic vocalization impairment of Foxp2 (R552H) knockin mice
related to speech-language disorder and abnormality of Purkinje cells. Proc. Natl.
Acad. Sci. U.S.A. 105, 31173122 (2008).
43. H. Huang, A. Yamamoto, M. A. Hossain, L. Younes, S. Mori, Quantitative cortical
mapping of fractional anisotropy in developing rat brains. J. Neurosci. 28, 14271433
44. W. Shu et al., Altered ultrasonic vocalization in mice with a disruption in the
Foxp2 gene. Proc. Natl. Acad. Sci. U.S.A. 102, 96439648 (2005).
45. E. D. Jarvis, Learned birdsong and the neurobiology of human language. Ann. N. Y.
Acad. Sci. 1016, 749777 (2004).
46. M. Kudo, M. S. Brem, W. M. Canfield, Mucolipidosis II (I-cell disease) and mucolipidosis
IIIA (classical pseudo-hurler polydystrophy) are caused by mutations in the GlcNAc-
phosphotransferase alpha / beta -subunits precursor gene. Am. J. Hum. Genet. 78,
451463 (2006).
47. R. A. Steet et al., A splicing mutation in the alpha/beta GlcNAc-1-phosphotransferase
gene results in an adult onset form of mucolipidosis III associated with sensory neu-
ropathy and cardiomyopathy. Am. J. Med. Genet. A. 132A, 369375 (2005).
48. M. Encarnação et al., Molecular analysis of the GNPTAB and GNPTG genes in 13 pa-
tients with mucolipidosis type II or type IIIIdentification of eight novel mutations.
Clin. Genet. 76,7684 (2009).
49. T. U. Han et al., A study of the role of the FOXP2 and CNTNAP2 genes in persistent
developmental stuttering. Neurobiol. Dis. 69,2331 (2014).
50. M. Pekny, M. Nilsson, Astrocyte activation and reactive gliosis. Glia 50, 427434
51. M. Bélanger, I. Allaman, P. J. Magistretti, Brain energy metabolism: Focus on
astrocyte-neuron metabolic cooperation. Cell Metab. 14, 724738 (2011).
52. M. Santello, A. Volterra, Synaptic modulation by astrocytes via Ca2+-dependent
glutamate release. Neuroscience 158, 253259 (2009).
53. R. Piet, L. Vargová, E. Syková, D. A. Poulain, S. H. Oliet, Physiological contribution of
the astrocytic environment of neurons to intersynaptic crosstalk. Proc. Natl. Acad. Sci.
U.S.A. 101, 21512155 (2004).
54. W. Walz, Role of astrocytes in the clearance of excess extracellular potassium. Neu-
rochem. Int. 36, 291300 (2000).
55. M. A. Anderson et al., Astrocyte scar formation aids central nervous system axon
regeneration. Nature 532, 195200 (2016).
56. J. J. Rodríguez-Arellano, V. Parpura, R. Zorec, A. Verkhratsky, Astrocytes in physio-
logical aging and Alzheimers disease. Neuroscience 323, 170182 (2016).
57. Y. Kikuchi et al., Abnormal auditory synchronization in stuttering: A magneto-
encephalographic study. Hear. Res. 344,8289 (2017).
58. Y. Zhang et al., An RNA-sequencing transcriptome and splicing database of glia,
neurons, and vascular cells of the cerebral cortex. J. Neurosci. 34, 1192911947 (2014).
59. Y. Zhang et al., Purification and characterization of progenitor and mature human
astrocytes reveals transcriptional and functional differences with mouse. Neuron 89,
3753 (2016).
60. M. Yang, M. D. Weber, J. N. Crawley, Light phase testing of social behaviors: Not a
problem. Front. Neurosci. 2, 186191 (2008).
61. W. C. Skarnes et al., A conditional knockout resource for the genome-wide study of
mouse gene function. Nature 474, 337342 (2011).
10 of 10
| Han et al.
Downloaded by guest on October 7, 2020
... Found altered vocalization patterns in the knock-in mouse carrying Gpnmb mutation. Han et al. [37] 2019 Characterized mouse model of stuttering with mutations in the Gnptab gene Detected altered ultrasonic vocalization in the knock-in mouse and found astrocyte deficits in the corpus callosum Choo et al. [40] 2011 Analyzed gene expression data at Allen Brain Institute and voxel-based morphometry Association of lysosomal enzyme trafficking genes with area size of gray matter in stutters and normal subjects SNP, single-nucleotide polymorphism; LOD, logarithm of odds ratio. ...
... While the mouse model of stuttering displayed abnormal features similar to human stutters, the location and role of the speech center, which is assumed to be in the brain, are entirely unknown. Han et al. [37] cleverly introduced Gnptab Ser321Gly and Ala455Ser mutations into mice by using a cell type-specific Cre-drivers and conditional knock-out tools, and reported that these mice showed altered ultrasonic vocalizations but normal in non-vocal behaviors. In addition, they found that only the astrocyte-specific Gnptab knockout mice showed an abnormal vocalization pattern, which raised the hypothesis that brain astrocytes, particularly in the corpus callosum, may be one of the potential brain regions where speech neurons may reside. ...
... Because of the critical regulatory role of astrocytes in the locomotion, mastication, and respiratory CPGs, by extrapolation, we propose that astrocytes residing within the vocalization CPG can also modulate activities of vocal production behaviors. Interestingly, astrocytes in other brain regions have been proposed to play a key role in the pathophysiology of a speech disorder, namely childhood-onset fluency disorder (stuttering) (Han et al., 2019;Maguire et al., 2021;Turk et al., 2021). Thus, it is plausible that a defect in function of astrocytes in the vocalization CPG may underlie the pathophysiological mechanisms for other forms of speech disorders. ...
Central pattern generators (CPGs) generate the rhythmic and coordinated neural features necessary for the proper conduction of complex behaviors. In particular, CPGs are crucial for complex motor behaviors such as locomotion, mastication, respiration, and vocal production. While the importance of these networks in modulating behavior is evident, the mechanisms driving these CPGs are still not fully understood. On the other hand, accumulating evidence suggests that astrocytes have a significant role in regulating the function of some of these CPGs. Here, we review the location, function, and role of astrocytes in locomotion, respiration, and mastication CPGs and propose that, similarly, astrocytes may also play a significant role in the vocalization CPG. • CPGs are crucial for complex motor behaviors such as locomotion, mastication, and respiration. Astrocytes have a significant circuit‐specific role in regulating the function of these CPGs. We propose that astrocytes may have a critical role in modulation of vocal production CPG.
... Lastly, given the link between stuttering-like disfluency and freezing behavior, one can speculate that genes influencing the prevalence of specific defense avoidance behaviors may influence developmental trajectories of stuttering. Mice with an engineered GNPTAB gene, which has been linked to stuttering in humans, exhibit deficits in the flow of ultrasonic vocalizations (Han et al., 2019). Animal models may shed light on episodic freezing during vocalization that contributes to stuttering-like disfluencies. ...
The purpose of this article is to provide a theoretical account of the experience of stuttering that incorporates previous explanations and recent experimental findings. According to this account, stuttering-like disfluencies emerge during early childhood from excessive detection of cognitive conflict due to subtle limitations in speech and language processes. For a subset of children who begin to stutter, the development of approach-avoidance motivational conflict likely contributes to a chronic reliance on cognitive control processes during speech. Consequently, maladaptive activation of right hemisphere inhibitory cortices to the basal ganglia via a hyperdirect pathway results in involuntary, episodic, and transient freezing of the motor system during speech initiation. This freeze response, consistent with defensive behavior in threatening situations, may lead to stuttering persistence, tension and struggle, maladaptive speech physiology, and feelings of anxiety and loss of control.
... Therefore, the therapeutic effects of methylphenidate may occur when it elicits slow, steady-state synaptic dopamine release [20]. In addition, recently it was proposed that astrocytes, the star-shaped non-neuronal cells in the brain, might have a key role in pathophysiology of stuttering [18,23,24]. The involvement of astrocytes in dopaminergic signaling and their modulatory role of neuronal circuits in the basal ganglia [18] further suggest that the underlying mechanisms of stuttering are not simple. ...
... 9,12,13 Follow-up studies have demonstrated that disruptions in GNPTAB resulted in deficits in astrocyte pathology in the corpus callosum and disruptions in mouse vocalization. 14 The roles of these astrocytes in the onset of stuttering are not well characterized; however, recent studies have contributed to a growing body of evidence that dopamine receptor D2 blockers can impact stuttering behavior, perhaps because of increased astrocyte metabolism in the striatum. 15 These studies suggest that dopamine projection from the basal ganglia might contribute to disturbances in speech and vocalization, and the findings potentially support pharmacological means for treatment. ...
Full-text available
Developmental stuttering is a speech disorder characterized by disruption in the forward movement of speech. This disruption includes part-word and single-syllable repetitions, prolongations, and involuntary tension that blocks syllables and words, and the disorder has a life-time prevalence of 6–12%. Within Vanderbilt’s electronic health record (EHR)-linked biorepository (BioVU), only 142 individuals out of 92,762 participants (0.15%) are identified with diagnostic ICD9/10 codes, suggesting a large portion of people who stutter do not have a record of diagnosis within the EHR. To identify individuals affected by stuttering within our EHR, we built a PheCode-driven Gini impurity-based classification and regression tree model, PheML, by using comorbidities enriched in individuals affected by stuttering as predicting features and imputing stuttering status as the outcome variable. Applying PheML in BioVU identified 9,239 genotyped affected individuals (a clinical prevalence of ∼10%) for downstream genetic analysis. Ancestry-stratified GWAS of PheML-imputed affected individuals and matched control individuals identified rs12613255, a variant near CYRIA on chromosome 2 (B = 0.323; p value = 1.31 × 10⁻⁸) in European-ancestry analysis and rs7837758 (B = 0.518; p value = 5.07 × 10⁻⁸), an intronic variant found within the ZMAT4 gene on chromosome 8, in African-ancestry analysis. Polygenic-risk prediction and concordance analysis in an independent clinically ascertained sample of developmental stuttering cases validate our GWAS findings in PheML-imputed affected and control individuals and demonstrate the clinical relevance of our population-based analysis for stuttering risk.
... 9,14,15,[50][51][52] Moreover, a recent study showed that knock-in mice carrying Gnptab mutations, homologous to previously identified human stuttering mutations and functionally related to AP4E1, exhibited reduced astrocyte density and volume in the corpus callosum together with vocalization deficits similar to those in human stuttering. 53 This animal study and our current study both indicate that structural abnormalities in the corpus callosum can be driven by specific genetic factors. However, the roles of the corpus collosum in speech production and stuttering are not fully understood. ...
Full-text available
Developmental stuttering is a common speech disorder with strong genetic underpinnings. Recently, stuttering has been associated with mutations in genes involved in lysosomal enzyme trafficking. However, how these mutations affect the brains of people who stutter remains largely unknown. In this study, we compared gray matter volume and white matter fractional anisotropy between a unique group of seven subjects who stutter and carry the same rare heterozygous AP4E1 coding mutations and seven unrelated controls without such variants. The carriers of the AP4E1 mutations are members of a large Cameroonian family in which the association between AP4E1 and persistent stuttering was previously identified. Compared to controls, mutation carriers showed reduced gray matter volume in the thalamus, visual areas and the posterior cingulate cortex. Moreover, reduced fractional anisotropy was observed in the corpus callosum, consistent with the results of previous neuroimaging studies of people who stutter with unknown genetic backgrounds. Analysis of gene expression data showed that these structural differences appeared at the locations in which expression of AP4E1 is relatively high. Moreover, the pattern of gray matter volume differences was significantly associated with AP4E1 expression across the left supratentorial regions. This spatial congruency further supports the connection between AP4E1 mutations and the observed structural differences. While gene mutations associated with stuttering have begun to be identified, their effects on human brains remain largely unknown. Chow et al. report that stuttering mutations in the AP4E1 gene are associated with structural differences in the corpus callosum and the thalamus, linking genetics and brain anomalies in stuttering.
Mucanje je neurorazvojni poremećaj složene etiologije, koja obuhvaća genetičke i okolinske komponente. Istraživanja pokazuju kako su u podlozi mucanja genetički faktori koji dovode do sklonosti mucanju, a ta sklonost se manifestira ako u okolini postoje umjereni okidači koji dovode do pojave mucanja (Rautakoski i sur., 2012). Prema istraživanjima, ako postoji mucanje unutar obiteljske anamneze, ono se može naslijediti - i s majčine i s očeve strane. Također, poznato je kako su dječaci skloniji razvoju mucanja u odnosu na djevojčice, indicirajući tako ulogu spola u početku, jakosti i nasljednosti mucanja. Gledajući samo obiteljsku anamnezu osobe koja muca, 70 % je moguće da je netko unutar obiteljskog stabla mucao. Kao dodatak tome idu u prilog teze iz istraživanja blizanaca. Istraživanje trojki pokazalo je višu razinu pojave mucanja kod jednojajčanih (20-90 %) u usporedbi s dvojajčanim blizancima (3-19 %) (WittkeThomson i sur., 2007). Utvrđena je povezanost 4 različita gena na nekoliko različitih kromosoma. Genetičke studije identificirale su varijante na genima GNPTAB, GNPTG, NAGPA i AP4E1, koje su povezane s pojavom mucanja (Devi i sur., 2021). Nadalje, dokazano je kako su neka perzistentna mucanja uzrokovana varijantama gena koji ne upravljaju govornim procesima, već metaboličkim putovima. Svi navedeni geni uključeni su u proces intracelularnog trgovanja, a deficiti u ovoj staničnoj funkciji su prepoznani kao uzročnici raznih neuroloških poremećaja (Frigerio-Domingues i sur., 2019). Neurološka perspektiva otkriva hiperaktivnost dopamina i abnormalnosti bijele tvari koje se uočavaju u mucanju, ukazujući na moguću neurokemijsku podlogu s još nejasnim mehanizmom djelovanja (Devi i sur., 2021).
Background Developmental stuttering is thought to be underpinned by structural impairments in the brain. The only way to support the claim that these are causal is to determine if they are present before onset. Materials and Methods Magnetic resonance imaging (MRI) was conducted on 18 neonates, aged 8–18 weeks, 6 of whom were determined to be genetically at risk of stuttering. Results With tract-based spatial statistics (TBSS) analysis, no statistically significant differences were found between the at-risk group and the control group. However, fractional anisotropy (FA), mean diffusivity (MD), and radial diffusivity (RD) in the corpus callosum of the at-risk group were lower (uncorrected) than in the control group. Automated Fiber Quantification (AFQ) yielded lower FA in the at-risk group than in the control group in the medial section of the callosum forceps minor. Discussion The findings, albeit with a small number of participants, support the proposition that reduced integrity of white matter in the corpus callosum has a causal role in developmental stuttering. Longitudinal research to determine if children with this impairment at birth later start to stutter is needed to confirm this. The left arcuate fasciculus is thought to develop as speech develops, which likely explains why there were no abnormal findings in this area in our at-risk neonates so soon after birth. This is the first study to investigate the brains of children before the onset of stuttering and the findings warrant further research.
Discovering developmental stuttering’s biological explanation has been an enduring concern. Novel advances in genomics and neuroscience are making it possible to isolate and pinpoint genetic and brain differences implicated in stuttering. This is giving rise to a hope that, in the future, dysfluency could be better managed if stuttering’s biological basis could be better understood. Concurrent to this, there is another hope rising: a hope of a future where differing fluencies would not be viewed through a reductive lens of biology and associated pathologies. The central aim of this paper is to edge out ethical implications of novel research into stuttering’s biological explanation. In doing so, the paper proposes to look beyond the bifurcation sketched by the medical and social model of disability. The paper demonstrates how the scientific hope of discovering stuttering’s biological explanation acts as an accessory of disablement due to the language of ‘lack’ and ‘deficit’ employed in reporting scientific findings and proposes participatory research with people who stutter as an antidote to manage this disablement.
Full-text available
Stuttering is a disorder characterized by intermittent loss of volitional control of speech movements. This hypothesis and theory article focuses on the proposal that stuttering may be related to an impairment of the energy supply to neurons. Findings from electroencephalography (EEG), brain imaging, genetics, and biochemistry are reviewed: (1) Analyses of the EEG spectra at rest have repeatedly reported reduced power in the beta band, which is compatible with indications of reduced metabolism. (2) Studies of the absolute level of regional cerebral blood flow (rCBF) show conflicting findings, with two studies reporting reduced rCBF in the frontal lobe, and two studies, based on a different method, reporting no group differences. This contradiction has not yet been resolved. (3) The pattern of reduction in the studies reporting reduced rCBF corresponds to the regional pattern of the glycolytic index (GI; Vaishnavi et al., 2010 ). High regional GI indicates high reliance on non-oxidative metabolism, i.e., glycolysis. (4) Variants of the gene ARNT2 have been associated with stuttering. This gene is primarily expressed in the brain, with a pattern roughly corresponding to the pattern of regional GI. A central function of the ARNT2 protein is to act as one part of a sensor system indicating low levels of oxygen in brain tissue and to activate appropriate responses, including activation of glycolysis. (5) It has been established that genes related to the functions of the lysosomes are implicated in some cases of stuttering. It is possible that these gene variants result in a reduced peak rate of energy supply to neurons. (6) Lastly, there are indications of interactions between the metabolic system and the dopamine system: for example, it is known that acute hypoxia results in an elevated tonic level of dopamine in the synapses. Will mild chronic limitations of energy supply also result in elevated levels of dopamine? The indications of such interaction effects suggest that the metabolic theory of stuttering should be explored in parallel with the exploration of the dopaminergic theory.
Full-text available
Evidence for genetic factors in persistent developmental stuttering has accumulated over the past four decades, and the genes that underlie this disorder are starting to be identified. The genes identified to date, all point to deficits in intracellular trafficking in this disorder.
Full-text available
Development of proficient spoken language skills is disrupted by mutations of the FOXP2 transcription factor. A heterozygous missense mutation in the KE family causes speech apraxia, involving difficulty producing words with complex learned sequences of syllables. Manipulations in songbirds have helped to elucidate the role of this gene in vocal learning, but findings in non-human mammals have been limited or inconclusive. Here, we performed a systematic study of ultrasonic vocalizations (USVs) of adult male mice carrying the KE family mutation. Using novel statistical tools, we found that Foxp2 heterozygous mice did not have detectable changes in USV syllable acoustic structure, but produced shorter sequences and did not shift to more complex syntax in social contexts where wildtype animals did. Heterozygous mice also displayed a shift in the position of their rudimentary laryngeal motor cortex (LMC) layer-5 neurons. Our findings indicate that although mouse USVs are mostly innate, the underlying contributions of FoxP2 to sequencing of vocalizations are conserved with humans.
Full-text available
The FOXP2 gene is important for the development of proper speech motor control in humans. However, the role of the gene in general vocal behavior in other mammals, including mice, is unclear. Here, we track the vocal development of Foxp2 heterozygous knockout (Foxp2+/−) mice and their wildtype (WT) littermates from juvenile to adult ages, and observe severe abnormalities in the courtship song of Foxp2+/− mice. In comparison to their WT littermates, Foxp2+/− mice vocalized less, produced shorter syllable sequences, and possessed an abnormal syllable inventory. In addition, Foxp2+/− song also exhibited irregular rhythmic structure, and its development did not follow the consistent trajectories observed in WT vocalizations. These results demonstrate that the Foxp2 gene is critical for normal vocal behavior in juvenile and adult mice, and that Foxp2 mutant mice may provide a tractable model system for the study of the gene’s role in general vocal motor control.
Full-text available
The functional and molecular similarities and distinctions between human and murine astrocytes are poorly understood. Here, we report the development of an immunopanning method to acutely purify astrocytes from fetal, juvenile, and adult human brains and to maintain these cells in serum-free cultures. We found that human astrocytes have abilities similar to those of murine astrocytes in promoting neuronal survival, inducing functional synapse formation, and engulfing synaptosomes. In contrast to existing observations in mice, we found that mature human astrocytes respond robustly to glutamate. Next, we performed RNA sequencing of healthy human astrocytes along with astrocytes from epileptic and tumor foci and compared these to human neurons, oligodendrocytes, microglia, and endothelial cells (available at With these profiles, we identified novel human-specific astrocyte genes and discovered a transcriptome-wide transformation between astrocyte precursor cells and mature post-mitotic astrocytes. These data represent some of the first cell-type-specific molecular profiles of the healthy and diseased human brain. Zhang et al. developed a method to acutely purify healthy and diseased human astrocytes and to culture them in serum-free conditions. They obtained transcriptome profiles of purified human CNS cell types and discovered two distinct stages of human astrocyte development.
Purpose: We combined a large longitudinal neuroimaging dataset that includes children who do and do not stutter and a whole-brain network analysis in order to examine the intra- and inter-network connectivity changes associated with stuttering. Additionally, we asked whether whole brain connectivity patterns observed at the initial year of scanning could predict persistent stuttering in later years. Methods: A total of 224 high-quality resting state fMRI scans collected from 84 children (42 stuttering, 42 controls) were entered into an independent component analysis (ICA), yielding a number of distinct network connectivity maps ("components") as well as expression scores for each component that quantified the degree to which it is expressed for each child. These expression scores were compared between stuttering and control groups' first scans. In a second analysis, we examined whether the components that were most predictive of stuttering status also predicted persistence in stuttering. Results: Stuttering status, as well as stuttering persistence, were associated with aberrant network connectivity involving the default mode network and its connectivity with attention, somatomotor, and frontoparietal networks. The results suggest developmental alterations in the balance of integration and segregation of large-scale neural networks that support proficient task performance including fluent speech motor control. Conclusions: This study supports the view that stuttering is a complex neurodevelopmental disorder and provides comprehensive brain network maps that substantiate past theories emphasizing the importance of considering situational, emotional, attentional and linguistic factors in explaining the basis for stuttering onset, persistence, and recovery.
In a previous magnetoencephalographic study, we showed both functional and structural reorganization of the right auditory cortex and impaired left auditory cortex function in people who stutter (PWS). In the present work, we reevaluated the same dataset to further investigate how the right and left auditory cortices interact to compensate for stuttering. We evaluated bilateral N100m latencies as well as indices of local and inter-hemispheric phase synchronization of the auditory cortices. The left N100m latency was significantly prolonged relative to the right N100m latency in PWS, while healthy control participants did not show any inter-hemispheric differences in latency. A phase-locking factor (PLF) analysis, which indicates the degree of local phase synchronization, demonstrated enhanced alpha-band synchrony in the right auditory area of PWS. A phase-locking value (PLV) analysis of inter-hemispheric synchronization demonstrated significant elevations in the beta band between the right and left auditory cortices in PWS. In addition, right PLF and PLVs were positively correlated with stuttering frequency in PWS. Taken together, our data suggest that increased right hemispheric local phase synchronization and increased inter-hemispheric phase synchronization are electrophysiological correlates of a compensatory mechanism for impaired left auditory processing in PWS.
A promising approach to understanding the mechanistic basis of speech is to study disorders that affect speech without compromising other cognitive or motor functions. Stuttering, also known as stammering, has been linked to mutations in the lysosomal enzyme-targeting pathway, but how this remarkably specific speech deficit arises from mutations in a family of general "cellular housekeeping" genes is unknown. To address this question, we asked whether a missense mutation associated with human stuttering causes vocal or other abnormalities in mice. We compared vocalizations from mice engineered to carry a mutation in the Gnptab (N-acetylglucosamine-1-phosphotransferase subunits alpha/beta) gene with wild-type littermates. We found significant differences in the vocalizations of pups with the human Gnptab stuttering mutation compared to littermate controls. Specifically, we found that mice with the mutation emitted fewer vocalizations per unit time and had longer pauses between vocalizations and that the entropy of the temporal sequence was significantly reduced. Furthermore, Gnptab missense mice were similar to wild-type mice on an extensive battery of non-vocal behaviors. We then used the same language-agnostic metrics for auditory signal analysis of human speech. We analyzed speech from people who stutter with mutations in this pathway and compared it to control speech and found abnormalities similar to those found in the mouse vocalizations. These data show that mutations in the lysosomal enzyme-targeting pathway produce highly specific effects in mouse pup vocalizations and establish the mouse as an attractive model for studying this disorder.
Transected axons fail to regrow in the mature central nervous system. Astrocytic scars are widely regarded as causal in this failure. Here, using three genetically targeted loss-of-function manipulations in adult mice, we show that preventing astrocyte scar formation, attenuating scar-forming astrocytes, or ablating chronic astrocytic scars all failed to result in spontaneous regrowth of transected corticospinal, sensory or serotonergic axons through severe spinal cord injury (SCI) lesions. By contrast, sustained local delivery via hydrogel depots of required axon-specific growth factors not present in SCI lesions, plus growth-activating priming injuries, stimulated robust, laminin-dependent sensory axon regrowth past scar-forming astrocytes and inhibitory molecules in SCI lesions. Preventing astrocytic scar formation significantly reduced this stimulated axon regrowth. RNA sequencing revealed that astrocytes and non-astrocyte cells in SCI lesions express multiple axon-growth-supporting molecules. Our findings show that contrary to the prevailing dogma, astrocyte scar formation aids rather than prevents central nervous system axon regeneration.
Recycling endosomes consist of a tubular network that emerges from vacuolar sorting endosomes and diverts cargoes toward the cell surface, the Golgi, or lysosome-related organelles. How recycling tubules are formed remains unknown. We show that recycling endosome biogenesis requires the protein complex BLOC-1. Mutations in BLOC-1 subunits underlie an inherited disorder characterized by albinism, the Hermansky-Pudlak Syndrome, and are associated with schizophrenia risk. We show here that BLOC-1 coordinates the kinesin KIF13A-dependent pulling of endosomal tubules along microtubules to the Annexin A2/actin-dependent stabilization and detachment of recycling tubules. These components cooperate to extend, stabilize and form tubular endosomal carriers that function in cargo recycling and in the biogenesis of pigment granules in melanocytic cells. By shaping recycling endosomal tubules, our data reveal that dysfunction of the BLOC-1-KIF13A-Annexin A2 molecular network underlies the pathophysiology of neurological and pigmentary disorders.