RhoE deficiency produces postnatal lethality, profound motor deficits and neurodevelopmental delay in mice

Abstract

Rnd proteins are a subfamily of Rho GTPases involved in the control of actin cytoskeleton dynamics and other cell functions such as motility, proliferation and survival. Unlike other members of the Rho family, Rnd proteins lack GTPase activity and therefore remain constitutively active. We have recently described that RhoE/Rnd3 is expressed in the Central Nervous System and that it has a role in promoting neurite formation. Despite their possible relevance during development, the role of Rnd proteins in vivo is not known. To get insight into the in vivo function of RhoE we have generated mice lacking RhoE expression by an exon trapping cassette. RhoE null mice (RhoE gt/gt) are smaller at birth, display growth retardation and early postnatal death since only half of RhoE gt/gt mice survive beyond postnatal day (PD) 15 and 100% are dead by PD 29. RhoE gt/gt mice show an abnormal body position with profound motor impairment and impaired performance in most neurobehavioral tests. Null mutant mice are hypoactive, show an immature locomotor pattern and display a significant delay in the appearance of the hindlimb mature responses. Moreover, they perform worse than the control littermates in the wire suspension, vertical climbing and clinging, righting reflex and negative geotaxis tests. Also, RhoE ablation results in a delay of neuromuscular maturation and in a reduction in the number of spinal motor neurons. Finally, RhoE gt/gt mice lack the common peroneal nerve and, consequently, show a complete atrophy of the target muscles. This is the first model to study the in vivo functions of a member of the Rnd subfamily of proteins, revealing the important role of Rnd3/RhoE in the normal development and suggesting the possible involvement of this protein in neurological disorders.

Figure 1. RhoE gene-trap strategy.

A) The gene trapping cassette in the retroviral vector VICTR37 was found in the second intron of the RhoE gene, as assessed by reverse PCR. LTR, viral long terminal repeat; SA, splice acceptor sequence; IRES, internal ribosome entry site; βGeo, fusion of beta-galactosidase and neomycin phosphotransferase genes; pA, polyadenylation sequence; PGK, phosphoglycerate kinase-1 promoter; BTK-SD, Bruton's tyrosine kinase splice donor sequence. B) Genotyping was performed as described in Materials and Methods by PCR. The position of the primers used is marked in A by arrows. The gene-trap allele (gt) yields a lower molecular weight band than the wild type allele (wt). C) Western blot with anti-RhoE antibodies to confirm the absence of RhoE expression in cells from RhoE gt/gt embryos. D) Sections showing RhoE immunoreactivity (RhoE-IR, left panel) and X-Gal staining (central and right panels) of a 14.5 dpc RhoE +/gt embryo to reveal expression of the RhoE locus. RhoE-IR is widespread in the embryo with a pattern that matches that of the X-Gal staining. RhoE expression is observed at the lumbar spinal cord (Sp c) and very intense in the striated muscles (Ms), whereas the primordium of the lumbar vertebral body (L v) lacks RhoE-IR (left panel) and X-Gal staining (central panel). A longitudinal section of the hindlimb of the same embryo (right panel) shows high level of X-Gal labeling at the level of the striated muscles and in the joints, whereas the bone primordia lack X-Gal staining. Ms: striated muscles; Tb: tibia primordium. The asterisk marks the knee joint primordium. Scale bar is 100 µm in the left and central panels and 200 µm in the panel on the right.

Figure 2. RhoE gt/gt mice are smaller in size and present a limited survival.

A) Pictures of RhoE gt/gt and +/+ littermates were taken at postnatal day (PD) 3 (left) and PD15 (right) to compare their size. Note the abnormal position of the hindlimbs of the RhoE gt/gt mouse. B) All mice (n = 49, 8 wt, 30 +/gt and 11 gt/gt) were weighted everyday from PD1 to PD21. Data are represented as Mean+SD. RhoE gt/gt mice were smaller than RhoE +/+ or +/gt animals at birth (Student's t test, p<0.01) and showed a growth delay thereafter (one way ANOVA and Tukey's test, p<0.005). C) Kaplan Meier survival curve of a total of 31 RhoE gt/gt mice. All RhoE +/+ and +/gt littermates survived beyond the day when the last RhoE gt/gt mouse died (not shown). The median survival for the RhoE gt/gt mice was 15 days.

Figure 3. RhoE gt/gt mice displayed neurobehavioral abnormalities.

The different reflexes and responses were analyzed as described in Materials and Methods. In all cases the postnatal day (PD) when the response appeared or was lost was recorded. A total of 49 mice (11 RhoE gt/gt, 30 +/gt and 8 +/+) were analyzed. Data are presented as Mean+SEM. Statistically significant differences between RhoE gt/gt and +/+ mice are shown (***p<0.001) A) Grasping reflexes and placing responses correlate with the altered hindlimb position in RhoE gt/gt mice. The day when the mice were able to show the correct response was recorded. B) Archaic reflexes (rooting and crossed extensor) persist longer in RhoE gt/gt mice than in wt or heterozygous littermates. Columns show the day when the reflex was not observed. C) The day when the following development landmarks appeared was recorded: separation of the ears from the head (Pinna), apparition of body hair (Hair) and opening of the auditive conduct (Ear) and of the eyelids (Eyes). The delay in the auditory (Preyer) and not in the tactile (Blast) startle responses is a functional consequence of the delayed ear opening.

Figure 4. RhoE gt/gt mice show neuromuscular development defects.

The same cohort of animals described in Figure 3 was used. Data are presented as Mean+SEM. Statistical differences between groups are shown when significantly different from the wt controls (*p<0.05; **p<0.01; ***p<0.001). A) The number of times that the mouse rotated on its hindlimbs without moving (Pivoting) over a 60 second period was recorded at different postnatal days (PD) as indicated. B) Latency to walk on a straight line was recorded. At PD 3 and 4, RhoE gt/gt mice showed no walking activity along the maximum time allowed (60 seconds). C) In the wire suspension test the latency to fall was significantly reduced in RhoE gt/gt mice. D) RhoE gt/gt mice perform worse in different motor tests. The postnatal day (PD) in which the following tests were performed correctly was recorded: vertical climbing (Climbing), vertical clinging (Clinging), righting reflex (Righting) and negative geotaxis (Geotaxis). The tests are described in Materials and Methods. E) Pictures show an example of the limb clasping response of a PD21 RhoE gt/gt mouse (right) compared with the normal escape posture of a RhoE +/+ (left) and a RhoE +/gt (central) mouse when suspended by the tail. F) Sections of P15 cerebella showing X-Gal staining only in the external granular cell layer (EGL, left panel) and the absence of gross abnormalities in the RhoE gt/gt sample (right panel) compared with the wild type (central panel). The sections in the central and right panels are stained with 1% cresyl fast violet solution.

Figure 5. RhoE gt/gt mice show a delay in neuromuscular junction maturation and a decrease in the number of spinal motoneurons.

A) Representative synaptic AChR cluster morphologies in the neuromuscular junctions of wild type (left) and RhoE gt/gt (right) mice at postnatal day 21, both in the triceps brachii (TB, top) and in the gastrocnemius (G, bottom) muscles. The picture of the RhoE gt/gt mouse corresponds to the M4 stage of neuromuscular junction, whereas the image of the wild type shows a more developed M5 form of synaptic cluster. B) Type of synaptic AChR clusters in PD21 RhoE gt/gt and wild type mice according to the status of maturation in the triceps brachii (left) and gastrocnemius (right) muscles. The forms of synaptic clusters that are present in the RhoE gt/gt mice are less developed than in the wild types. C) Motoneurons from 4 RhoE wt and 5 RhoE gt/gt mice were counted as described in Materials and Methods. RhoE gt/gt mice show a significantly reduced number of motoneurons (1467±146 vs 2582±322, p<0.001 in a Student's t test). D) Representative pictures of RhoE +/+ and RhoE gt/gt ventral horns (dotted lines) of cervical spinal cord sections. Arrows point at one motoneuron in each section.

Figure 6. Abnormal walking of RhoE gt/gt mice.

Representative walking footprint patterns of PD21 wild type (top) and RhoE gt/gt (bottom) mice. Forepaws were stained in red and hindpaws in blue. The pattern clearly differs, showing shorter and irregularly spaced strides in RhoE gt/gt mice when compared to the wild types. The graph shows the quantification of the stride length index of wild type and RhoE gt/gt mice. Since RhoE gt/gt mice were smaller than the wild types, the stride length index was calculated as the ratio stride-length/body-length for each mouse. The stride length index was reduced in the RhoE gt/gt mice compared to the control mice (*p<0.05 in a Student's t test).

Figure 7. Absence of the common peroneal nerve and disappearance of the craniolateral hindlimb muscle fibers in RhoE gt/gt mice.

A) Top panels: Dissection of the hindlimb of representative PD21 wild type, RhoE +/gt and RhoE gt/gt mice. The common peroneal nerve was absent in all the RhoE gt/gt mice (right panel). Arrows indicate the presence of the common peroneal nerve in the wild type (left panel) and in the heterozygous (central panel) and its absence in the RhoE gt/gt sample (right panel). Botom panels: The two main components (tibial and common peroneal) of a sciatic nerve removed from a wild type mouse (left panel) and form a heterozygous central panel) can be observed, whereas the sciatic nerve of a RhoE gt/gt mouse only shows the tibial nerve. B) The spinal roots originating the sciatic nerve in a wild type (top) and in a RhoE gt/gt (bottom) mouse are similar. The figure on the right shows a schematic representation of the roots. C) Histological section of a left P15 hindlimb from a wild type (left panels) and a RhoE gt/gt (right panels) mouse. The lower panels show a higher magnification of the area in the upper rectangle. The dotted lines represent the area covered by the craniolateral muscles in the wild type and in the RhoE gt/gt mice. Note that in the RhoE gt/gt hindlimb almost all muscle cells have been replaced by adipose tissue and clusters of small size fibers. TA: tibialis anterior muscle, EDL: extensor digitorum longus muscle, T: tibia, F: fibula.