The Pafah1b Complex Interacts with the Reelin Receptor
Guangcheng Zhang1,2., Amir H. Assadi1,2., Robert S. McNeil1,2, Uwe Beffert7, Anthony Wynshaw-Boris8, Joachim Herz7, Gary D. Clark1,2,3,4.,
1The Cain Foundation Laboratories, Texas Children’s Hospital, Houston, Texas, United States of America, 2Department of Pediatrics, Baylor College of
Medicine, Houston, Texas, United States of America, 3Department of Neuroscience, Baylor College of Medicine, Houston, Texas, United States of
America, 4Department of Neurology, Baylor College of Medicine, Houston, Texas, United States of America, 5Program in Developmental Biology,
Baylor College of Medicine, Houston, Texas, United States of America, 6Program in Translational Biology and Molecular Medicine, Baylor College of
Medicine, Houston, Texas, United States of America, 7Department of Molecular Genetics, University of Texas Southwestern Medical Center, Dallas,
Texas, United States of America, 8Department of Pediatrics and Medicine, University of California, San Diego School of Medicine, La Jolla, California,
United States of America
Reelin is an extracellular protein that directs the organization of cortical structures of the brain through the activation of two
receptors, the very low-density lipoprotein receptor (VLDLR) and the apolipoprotein E receptor 2 (ApoER2), and the
phosphorylation of Disabled-1 (Dab1). Lis1, the product of the Pafah1b1 gene, is a component of the brain platelet-activating
factor acetylhydrolase 1b (Pafah1b) complex, and binds to phosphorylated Dab1 in response to Reelin. Here we investigated
the involvement of the whole Pafah1b complex in Reelin signaling and cortical layer formation and found that catalytic
subunits of the Pafah1b complex, Pafah1b2 and Pafah1b3, specifically bind to the NPxYL sequence of VLDLR, but not to
ApoER2. Compound Pafah1b1+/2;Apoer22/2mutant mice exhibit a reeler-like phenotype in the forebrain consisting of the
inversion of cortical layers and hippocampal disorganization, whereas double Pafah1b1+/2;Vldlr2/2mutants do not. These
results suggest that a cross-talk between the Pafah1b complex and Reelin occurs downstream of the VLDLR receptor.
Citation: Zhang G, Assadi AH, McNeil RS, Beffert U, Wynshaw-Boris A, et al (2007) The Pafah1b Complex Interacts with the Reelin Receptor
VLDLR. PLoS ONE 2(2): e252. doi:10.1371/journal.pone.0000252
Heterozygous mutations in the PAFAH1B1 (LIS1) gene in humans
causes a reduction in the number of cortical gyri (lissencephaly)
. Homozygous mutations in the REELIN (RELN) gene also
result in lissencephaly, with additional cerebellar hypoplasia .
Reelin, a secreted glycoprotein controlling neuronal positioning,
functions by clustering its receptors VLDLR and ApoER2, causing
the activation of src-family kinases (SRKs) and the phosphoryla-
tion of the adapter molecule Dab1 (reviewed by [3–5]). Disruption
of Reelin (Reln) in homozygous reeler mice results in cortical layer
disruption, cerebellar hypoplasia and ataxia. Mice deficient for
Dab1 [6–8], both VLDLR and ApoER2 receptors  and SFKs
Fyn and Src  exhibit a reeler-like phenotype.
Homozygous deletions of the Pafah1b1 gene (encoding Lis1) in
the mouse result in early embryonic lethality, whereas heterozy-
gous mutations lead to hippocampal lamination defects .
Further reduction of Lis1 activity in compound hypomorphic
mutants led to the disruption of cortical layers . Lis1 was
initially identified as the non-catalytic b subunit of the Pafah1b
complex [1,13]. This complex also contains two catalytic a subunits
encoded by the Pafah1b2 and Pafah1b3 genes that hydrolyze the
platelet-activating factor . The product of the Pafah1b2 gene
(a2) is 30 kDa, whereas the product of the Pafah1b3 gene (a1) is a 29
kDa protein. The entire Pafah1b complex resembles a G-protein
signaling complex [15,16]. In the mouse, mutations in the
a subunit genes cause no overt neurological phenotype, but loss
of Pafah1b2 disrupts spermatogenesis [16,17]. In humans,
PAFAH1B3 hemizygousity is associated with mental retardation
and ataxia , suggesting that the catalytic subunits of Pafah1b
may be important for brain development or function. In addition
to its involvement with the Pafah1b complex, Lis1 participates
in cytoskeletal dynamics as a component of an evolutionary
conserved pathway that mediates nucleokinesis (reviewed by ).
Lis1 regulates the function of cytoplasmic dynein/dynactin motor
complex [20,21] through binding to several of its components.
These interactions are thought to be important for several aspects
of brain development, including neural stem cell proliferation and
neuronal migration .
We previously showed that Lis1 binds Dab1 in a Reelin-
dependent manner and that Lis1 and Reelin functionally interact
. Given the association of Lis1 with the Pafah1b a subunits, we
investigated the interaction between this complex and the Reelin
pathway. Here we demonstrated that both a subunits bind the
Reelin receptor VLDLR but not ApoER2. Genetic experiments
further demonstrated that loss of ApoER2 combined with Lis1
reduction results in a reeler-like phenotype, suggesting that the
Pafah1b complex modulates the Reelin pathway.
Academic Editor: Ulrich Mueller, University of Giessen, Germany
Received January 16, 2007; Accepted January 30, 2007; Published February 28,
Copyright: ? 2007 Zhang et al. This is an open-access article distributed under
the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the
original author and source are credited.
Funding: Supported by a grant from the National Institute of Disorders and
Stroke/NIH R01 NS042616 (G.D.), NIH grants HL20948, HL63762, NS43408, the
Perot Family Foundation and the Humboldt Foundation (J.H.) and by the Mental
Retardation and Developmental Disabilities Research Center (MRDDRC) at Baylor
College of Medicine.
Competing Interests: The authors have declared that no competing interests
* To whom correspondence should be addressed. E-mail: email@example.com.
. These authors contributed equally to this work. (GZ and AA; GC and GD)
PLoS ONE | www.plosone.org1February 2007 | Issue 2 | e252
Pafah1b3 and Pafah1b2 bind to VLDLR but not to
Previous observations revealed the genetic interaction of Pafah1b1
with the genes encoding components of the Reln signaling path-
way, and the direct binding of Lis1 to phosphorylated Dab1 .
In this study we turned our attention to the catalytic a subunits
of the Pafah1b complex. To investigate potential biochemical
interactions of Pafah1b a subunits with the Reelin receptors
VLDLR and ApoER2, co-immunoprecipitation assays were per-
formed. Plasmids encoding Pafah1b3 (tagged with a MYC epitope)
and VLDLR (tagged with either GFP or HA epitopes) were
transfected in 293T cells. Results show that VLDLR was co-
immunoprecipitated with antibodies directed against the Pafah1b3
tag and, conversely, that Pafah1b3 was co-immunoprecipitated
with antibodies directed against the VLDLR tag (Fig. 1A).
To determine whether this binding was direct, individual
Figure 1. Pafah1b2 and Pafah1b3 bind VLDLR but not ApoER2. (A) Pafah1b3 binds VLDLR in transfected cells. Pafah1b3-MYC was co-expressed in
293T cells with GFP- (lanes 1–3) or HA-tagged VLDLR (lanes 4–6). VLDLR was co-precipitated with MYC antibodies directed against Pafah1b3 (lanes 2
and 5), but not with control antibodies (lanes 3 and 6). Conversely, Pafah1b3 was co-precipitated with HA (lane 9) or GFP (lane 10) directed against
VLDLR, but not with control antibodies (lanes 11–12). Lanes 1 and 4 show VLDLR, and lanes 7–8 show Pafah1b3 in the WCL. Blots were probed with
the GFP (lanes 1–3), HA (lanes 4–6) or Myc antibodies (lanes 7–12). (B) Pafah1b2 and Pafah1b3 bind VLDLR in a cell-free-system (TNT). In vitro
translated proteins were radiolabeled with35S and analyzed by SDS-PAGE (lanes 1–4). Individual Pafah1b subunits were combined with equal
amounts of VLDLR and immunoprecipitated with FLAG (lanes 5–7) or a negative control antibody (lanes 8). Proteins were detected by
autoradiography. Note that VLDLR co-precipitated with Pafah1b a subunits, but not with Lis1. (C) Pafah1b2 and Pafah1b3 do not bind ApoER2 in
transfected cells. ApoER2-GFP was co-expressed in 293T cells with the indicated proteins, and co-immunoprecipitated with HA antibodies directed
against Dab1 (lane 1). FLAG antibodies directed against the Pafah1b subunits (lanes 2–4) or a control antibody (lane 5) did not co-precipitate the
receptor. Blots were probed with antibodies against GFP to detect co-precipitated ApoER2 (upper panel) or total ApoER2 expression in the
corresponding WCLs (middle panel). FLAG or HA antibodies were used to detect Pafah1b subunits (lower left panel) or Dab1 (lower right panel) in the
corresponding WCLs. ApoER2 co-precipitated exclusively with Dab1. (D) Pafah1b2 and Pafah1b3 do not bind ApoER2 in a cell-free-system (TNT). In
vitro translated proteins were radiolabeled with35S and analyzed by SDS-PAGE (lanes 1–4). Equal amounts of ApoER2 were combined with the
indicated proteins and immunoprecipitated with a control antibody (lane 5), or with antibodies against Dab1 (lane 6), Pafah1b2 (lane 7) or Pafah1b3
(lane 8). Proteins were detected by autoradiography. ApoER2 again co-precipitated only with Dab1. IP, immunoprecipitate; WCL, whole cell lysate.
Pafah1b Binds Vldlr
PLoS ONE | www.plosone.org2February 2007 | Issue 2 | e252
FLAG-tagged Pafah1b subunits were expressed in a cell-free
system and incubated in the presence of in vitro- translated
VLDLR. The receptor was efficiently co-immunoprecipitated with
either Pafah1b2 or Pafah1b3, but not with Lis1 (Fig. 1B),
indicating that the a subunits are capable of binding VLDLR
Similar experiments were performed to determine whether
Pafah1b3 and Pafah1b2 interact with ApoER2. When this
receptor was co-expressed with either Pafah1b2 or Pafah1b3 in
293T cells, immunoprecipitation experiments failed to reveal any
interaction (Fig. 1C). As a positive control we used Dab1, which is
known to bind lipoprotein receptors . Immunoprecipitation
assays using in vitro translated proteins also revealed no interaction
between a subunits and ApoER2, whereas Dab1 was able to bind
the receptor, as expected (Fig. 1D). Together, these results
demonstrate that both Pafah1b2 and Pafah1b3 a subunits interact
specifically with the Reelin receptor VLDLR, but not with
The cytoplasmic NPxY domain of lipoprotein receptors enables
binding of proteins containing PTB domains such as Dab1 . To
determine whether the NPxY motif is required for Pafah1b2 and
Pafah1b3 binding to VLDLR, a series of C terminal truncation
mutants of the receptor tagged with GFP were produced (Fig. 2A).
Co-immunoprecipitation experiments were conducted in COS7
cells using antibodies against the a subunit tag. The VLDLR
ectodomain fragment (VLDLRD809) and the C-terminal trunca-
tion mutant VLDLRD825, both lacking the NPxY domain,
showed no interaction with Pafah1b3 (Fig. 2B) or with Pafah1b2
(Fig. 2D). On the other hand, the truncation construct
VLDLRD855 that retained the NPxY motif did exhibit binding
to Pafah1b3 (Fig. 2B and C) and to Pafah1b2 (Fig. 2D). Finally, the
specific NPxY mutant, VLDLR(AAxA), showed loss of interaction
with either Pafah1b3 (Fig. 2C) and with Pafah1b2 (Fig. 2D). These
results demonstrated that a subunit binding to VLDLR requires
the presence of an intact NPxY motif.
The cytoplasmic regions of VLDLR and ApoER2 that include
the NPxY binding domain possess significant sequence homology,
yet we found that Pafah1b2 and Pafah1b3 bind only to VLDLR.
To understand this apparent discrepancy, we examined the protein
sequence of the receptor intracellular domains. The NPxY motif
sequence in both receptors is NPVY, but the first amino acid
downstream of this motif differs. In VLDLR this residues corre-
sponds to a leucine (Leu838), whereas in ApoER2 is an arginine
(Arg774) (Fig. 3A). To determine whether a leucine at this position
is important for Pafah1b2 and Pafah1b3 binding, an ApoER2
mutant receptor was generated in which Arg774was substituted by
a leucine (R774L). Co-immunoprecipitation experiments in trans-
fected cells demonstrated that ApoER2 carrying the R774L
mutation was able to bind Pafah1b2 and Pafah1b3, unlike the wild
type receptor (Fig. 3B). These observations suggest that the
NPVYL sequence of VLDLR mediates its unique property of
binding to the Pafah1b a subunits. In contrast, Dab1 does not
appear to discriminate between lipoprotein receptors as it binds to
both, the NPVYL sequence of VLDLR and the NPVYR sequence
Since the Pafah1b a subunits and Dab1 bind a similar
region of VLDLR, we reasoned that they might compete
for receptor occupancy. Indeed, increasing concentrations of
Pafah1b2 were found to displace Pafah1b3 from VLDLR in co-
immunoprecipitation experiments in vitro (Fig. 4A). Furthermore,
increasing concentrations of Pafah1b3 were able to partially
displace Dab1 from VLDLR, consistently with our observation
that they bind similar residues of the receptor intracellular domain
Figure 2. Pafah1b2 and Pafah1b3 interact specifically with the NPxY
motif of VLDLR. (A) Diagram of the VLDLR expression constructs used in
this study. TM=transmembrane region. CT=cytoplasmic tail. (B) The
NPxY motif is required for VLDLR binding to Pafah1b3. Pafah1b3-MYC
was co-expressed in 293T cells with the indicated GFP-tagged VLDLR
constructs. Proteins were immunoprecipitated with Myc antibodies and
the blot was probed with GFP antibodies to detect VLDLR receptors
(upper panel). Only full-length VLDLR and the NPxY-contain-
ingVLDLR?855 co-precipitated with Pafah1b3. WCLs were probed with
GFP (middle panels) or Myc (lower panels) antibodies to ensure that
similar amounts of VLDLR or Pafah1b3 proteins were present in each
sample. (C) The NPxY motif is required for VLDLR binding to Pafah1b2.
FLAG-Pafah1b2 was co-expressed in 293T cells with GFP-tagged VLDLR
constructs and immunoprecipitated with FLAG antibodies. Blots were
probed with GFP antibodies to detect VLDLR proteins in the IP (upper
panel) or WCLs (middle panel), or FLAG antibodies to detect Pafah1b2
in the WCLs (lower panel). Only full-length VLDLR and VLDLR?855 co-
precipitated with Pafah1b2. IP, immunoprecipitate; WCL, whole cell
Pafah1b Binds Vldlr
PLoS ONE | www.plosone.org3February 2007 | Issue 2 | e252
In vivo analysis of the Pafah1b complex function in
The biochemical data presented above suggests that the a subunits
may bring the whole Pafah1b complex in proximity of the
VLDLR receptor. Since this receptor also binds Dab1, our
findings raised the possibility that the Pafah1b complex may
modulate Reelin signaling. To address this possibility in vivo, we set
out to generate double mutant mice carrying mutations in genes
encoding Pafah1b subunits as well as in genes encoding Reelin
receptors. Given that both, Apoer2 and Pafah1b2 male knock out
mice are sterile [16,24], we could not generate double mutants
that lacked these proteins. We were however able to readily
generate compound mice that were heterozygous for Pafah1b1 and
homozygous for either Apoer2 or Vldlr. Cortical lamination in these
mutants was analyzed as a read-out of Reelin activity in brain
development and compared to single mutants or to double Apoer2/
Vldlr mutants, which exhibit a reeler-like phenotype. Cortical
sections were stained with two cellular layer-specific neuronal
markers, Calbindin (layer II/III) and Foxp2 (layer VI) and the
layer distribution of immunolabeled cells was analyzed quantita-
tively. This analysis revealed no obvious lamination defects in
single Pafah1b1+/2or Vldlr2/2mice, whereas some layer abnor-
malities were observed in single Apoer22/2mice, as previously
reported [9,25] (Fig. 5). The cortex of Pafah1b1+/2;Vldlr2/2
double mutants also appear fairly normal, however, that of
Pafah1b1+/2;Apoer22/2double mutants presented a severe abnor-
mality similar to the cortical layer inversion typically seen in Reln2/
2, Dab12/2or Apoer22/2;Vldlr2/2mutants (Fig. 5). Calbindin-
positive neurons destined for upper layers were ectopically located
in the lower cortex, whereas Foxp2-positive neurons destined for
a lower layer were ectopically located into the upper cortex. The
cortex of Pafah1b1+/2;Apoer22/2mutants also revealed hypercel-
lularity of layer I, another typical feature of the reeler phenotype
that is also seen in double Apoer22/2;Vldlr2/2mutants (Fig. 5).
To gain further evidence of the occurrence of a reeler-like
phenotype in Pafah1b1+/2;Apoer22/2double mutants, we also
examined the anatomy of hippocampal structures (Fig. 6). Cellular
layers in the hippocampus proper and dentate gyrus were normal
in heterozygous Apoer2+/2mice, whereas a modest split of the
pyramidal layer in area CA1 and CA3 was observed in
Pafah1b1+/2and homozygous Apoer22/2mice. However, a reeler-
like phenotype characterized by profound dyslamination of all
cellular layers was observed in double Pafah1b1+/2;Apoer22/2mice
(Fig. 6). No gross abnormalities were observed in the cerebellum of
Pafah1b1+/2;Apoer22/2double mutants, unlike reeler mice which
exhibit profound cerebellar hypoplasia (not shown). Together,
these data demonstrate that Pafah1b1, like Vldlr mutations,
synergize with Apoer2 disruption and contribute to the appearance
of a reeler-like phenotype, at least in the forebrain.
Reelin signaling is largely intact in Pafah1b1+/2;
In normal neurons, Reelin treatment induces Dab1 phosphory-
lation on tyrosine residues [26–28] and the activation of PI3K,
which leads to the phosphorylation of Akt on serine residue 473
[29–31]. In double Vldlr2/2;Apoer22/2mutant mice the appear-
ance of the anatomical phenotype correlates with loss of these
Reelin-dependent signaling events . Since double Pafah1b1+/2;
Figure 3. Substitution of Arg774with a Leu in ApoER2 rescues binding to Pafah1b a subunits. (A) Amino acid sequence of VLDLR and ApoER2 near the
NPxY motif in the cytoplasmic tail of each receptor. Unique amino acids R774 in ApoER2 and L838 in VLDLR are underlined. (B) A Leu residue
following the NPxY motif is required for Pafah1b a subunit binding to lipoprotein receptors. FLAG-tagged intact ApoER2(WT) or a mutant receptor in
which Arg774was substituted by a Leu ApoER2(R774L) were co-expressed in COS7 cells with GFP-tagged Pafah1b a subunits. Proteins were
immunoprecipitated with FLAG antibodies (upper panel) and immunoblotted with antibodies against GFP to detect co-precipitated a subunits. FLAG
(middle panel) or GFP (lower panel) antibodies were used to detect ApoER2 receptors or the Pafah1b a subunits in the WCL. Pafah1b2 and Pafah1b2
co-precipitated with ApoER2(R774L), but not with intact ApoER2(WT). IP, immunoprecipitate; WCL, whole cell lysate.
Pafah1b Binds Vldlr
PLoS ONE | www.plosone.org4February 2007 | Issue 2 | e252
Apoer22/2mutants exhibit a reeler-like phenotype in the forebrain,
we sought to determine whether Reelin signaling was also affected.
Cortical neurons were obtained from mice carrying Pafah1b1,
Apoer2 and Vldlr mutations, alone or in combination. As for normal
mice, Reelin treatment was found to induce Dab1 and Akt
phosphorylation in all mutants examined, including Pafah1b1+/
2;Apoer22/2mutants, even though these animals exhibit cortical
layers defects (Fig. 7). These data indicate Dab1 and Akt
phosphorylation is not sufficient to induce cortical layer formation.
We have previously demonstrated the existence of an interaction
between Reelin and Lis1 signaling consisting of the direct binding
of Lis1, the regulatory subunit of the Pafah1b complex, to the
adapter Dab1 . This interaction takes place when Dab1 is
phosphorylated on tyrosine residues in response to Reelin. In the
present study we have examined the interaction of individual
Reelin receptors with the subunits of the Pafah1b complex. We
shown that the catalytic a subunits of the Pafah1b complex,
Pafah1b2 and Pafah1b3, bind specifically VLDLR and that
a reduction in Lis1 activity mimics the loss of this receptor in the
forebrain. The binding of Pafah1b3 and Pafah1b2 to VLDLR
requires the NPxY domain and the presence of a leucine residue
immediately following this sequence. The catalytic a subunits
cannot bind the NPxYR sequence of ApoER2, but a point
mutation that converts the arginine residue adjacent to the NPxY
motif to a leucine rescued Pafah1b a subunit binding, demon-
strating that this residue is critical for coupling the Pafah1b
complex selectively with VLDLR. Given the low abundance of
this receptor in neurons, we could not confirm that the interactions
we observed in transfected cells and in vitro also take place in
normal neurons. However, given the specificity of the Pafah1b
a subunits for VLDLR and the strict requirement for the NPVYL
sequence, it seems reasonable to conclude that the binding may
indeed occur in vivo.
Through our genetic studies, we demonstrated that the
biochemical interaction of the Pafah1b complex with VLDLR
has physiological consequences for forebrain development. Con-
sistent with our biochemical data, we observed that Pafah1b1+/2
mutations had no effect on the appearance of brain structures in
Vldlr2/2mutants, suggesting that the products of these genes may
function in a linear pathway. On the other hand Pafah1b1+/2
mutations exacerbated the phenotype of Apoer22/2mutants to an
extent that the appearance of cortical and hippocampal structures
in double Pafah1b1+/2; Apoer22/2mutants resembled that of reeler
mice. Since a reeler-like phenotype is also observed in double
Vldlr2/2; Apoer22/2mutants , these data suggest that Lis1 is an
important component of the Reelin signaling pathway down-
stream of VLDLR (Fig. 8). The simplest interpretation of our data
is that the a subunits function as signaling adapter molecules by
bringing Lis1 in proximity of the VLDLR receptor and Dab1, thus
facilitating Reelin signaling through this receptor. An alternative
interpretation of our genetic findings is that ApoER2 is the
dominant Reelin receptor in forebrain development, and that the
consequences of Pafah1b1+/2mutations on Reelin signaling can
only be appreciated when this receptor is missing. This view is
Figure 4. Pafah1b a subunits compete with each other and with Dab1 for binding to VLDLR. (A) Pafah1b a subunits compete for VLDLR binding.
Proteins were expressed in vitro using a cell-free-system (TNT) and radiolabeled with35S (lanes 1–3). Equal amounts (50 ml) of VLDLR and Pafah1b3
were incubated with increasing amounts of Pafah1b2, as indicated. Proteins were immunoprecipitated with polyclonal antibodies against VLDLR
(lanes 5–9) or a control antibody (lane 4) and detected by autoradiography. The plot represents the mean ratio of co-precipitated Pafah1b3 and
Pafah1b2 normalized to the amount of precipitated VLDLR in each sample. Increasing amounts of Pafah1b2 reduced the amount of co-precipitated
Pafah1b3. (B) Proteins were expressed in vitro (TNT) and radiolabeled with35S (lanes 1–3). Equal amounts (50 ml) of VLDLR and Dab1 were incubated
with increasing amounts of Pafah1b3, as indicated. Proteins were immunoprecipitated with polyclonal antibodies against VLDLR (lanes 5–9) or
a control antibody (lane 4) and detected by autoradiography. The plot represents the mean ratio of co-precipitated Pafah1b3 and Dab1 normalized to
the amount of precipitated VLDLR in each sample. Increasing amounts of Pafah1b3 reduced the amount of co-precipitated Dab1. Bars represent the
standard error of the mean from triplicate experiments. IP, immunoprecipitate.
Pafah1b Binds Vldlr
PLoS ONE | www.plosone.org5 February 2007 | Issue 2 | e252
supported by the observation that Apoer22/2mutations in isolation
already result in a noticeable cortical and hippocampal phenotype,
unlike Vldlr2/2mutations [9,25]. Both interpretations of our data
are consistent with a functional role for the Pafah1b complex in
Reelin signaling during brain development.
Despite the disorganization of cortical layers, we observed that
the induction of Dab1 and Akt phosphorylation by Reelin was
fairly normal in neurons isolated from Pafah1b1+/2; Apoer22/2
double mutant brains. This is in striking contrast to the results
obtained using Vldlr2/2; Apoer22/2double mutant neurons, in
which Reelin does not appear capable to elicit any signaling events
[25,29]. Similarly, Reelin treatment did not induce Akt phos-
phorylation in Dab12/2neurons [29,31]. Our findings indicate
that the Pafah1b complex and Lis1 are not required for many of
the signaling events which are normally stimulated by Reelin
mainly through clustering of the ApoER2 receptor. In the absence
of ApoER2, Reelin signaling events such as Dab1 and Akt
phosphorylation still occur, albeit to a lower level, mediated by the
VLDLR receptor and irrespective of the presence of Pafah1b
proteins. However, under these reduced signaling conditions, Lis1
deficiency prevents the formation of cortical layers. Together with
our previous observation that Lis1 binds exclusively phospho-
Dab1 , the present findings suggest that Lis1 functions
downstream of SFK activity and it is not predicted to interfere
with the interaction between Dab1 and other signaling molecules
such as PI3K [29–31], Nckb , Crk family proteins [28,33], or
We and others have previously shown that loss of Pafah1b
a subunits in the mouse does not result in a neurological pheno-
type, but affects spermatogenesis [16,17]. These studies indicate
that the catalytic subunits of the Pafah1b complex are not
absolutely required for brain development. Based on the present
Figure 5. reeler-like disruption of cortical layers in Pafah1b1+/2mice lacking ApoER2 but not VLDLR. Sagittal sections of the neocortex were obtained
from adult mice of the indicated genotype. Adjacent sections were stained with cresyl violet (CV) (a-f) or subjected to immunohistochemistry with
antibodies against calbindin to label cells in layers II–III (g-l), or Foxp2 to label cells in layer VI (s-x). Histograms represent the radial distribution of cells
positive for calbindin (m-r) or Foxp2 (y-dd) from the bottom of the cortical plate (set as 0) to the pial surface (set as 100). Pafah1b1+/2and Vldlr2/2
single or double mutants presented no obvious cortical layering defects. Apoer22/2single mutants exhibited some laminar dispersion of upper layer
neurons. In contrast, Pafah1b1+/2;Apoer22/2double mutants displayed a marked inversions of upper and lower layers (reeler-like phenotype)
comparable to that seen in Apoer22/2;Vldlr2/2mice. Scale bars=500 mm.
Pafah1b Binds Vldlr
PLoS ONE | www.plosone.org6 February 2007 | Issue 2 | e252
data we propose that they may modulate Reelin signaling
downstream of VLDLR, possibly by promoting Lis1 and Dab1
VLDLR and ApoER2 are both individually capable of binding
Reelin on the extracellular side and Dab1 on the intracellular side,
and both contribute to cortical layer formation [9,25,35,36]. In
addition, ApoER2 is known to bind JNK Interacting Protein (JIP)
1 and 2 and PSD-95 through a unique intracellular domain
encoded by an alternatively spliced exon [37–39]. Recent studies
demonstrated that ApoER2 interacts with the NMDA receptor,
thereby mediating a Reelin-dependent function in learning and
memory in the adult brain . Here we have shown that
VLDLR is also capable of unique interactions that may affect
Reelin signaling and forebrain cellular layer formation. It has
recently been reported that VLDLR deletions in humans result in a
neurological disorder characterized by lissencephaly and cerebel-
lar hypoplasia, malformations similar but less severe than those
associated with RELN deletions . Thus, in humans an overt
neurological phenotype is seen even in the presence of the
ApoER2, further underscoring the importance of VLDLR in brain
development. It remains to be determined whether the unique
ability of VLDLR to bind the Pafah1b complex affects postnatal
brain function in addition to neuronal positioning during
MATERIALS AND METHODS
Generation of plasmid constructs
Human VLDLR cDNA encoding the 873 amino acids isoform A
(accession # NP_003374) was cloned into pEGFP-N1 (Clontech)
to generate the VLDLR-GFP fusion protein. Alternatively, the
same cDNA was tagged at the C terminus with the HA epitope by
PCR. Truncation constructs VLDLRD809-GFP (containing
VLDLR amino acids 1-809), VLDLRD825-GFP (containing
VLDLR amino acids 1-825) and VLDLRD855-GFP (containing
VLDLR amino acids 1-855) were generated using the Expand
High Fidelity PCR System. To produce VLDLR(AAxA)-GFP, site
directed mutagenesis was performed on the VLDLR-GFP plasmid
using the Stratagene QuickChange Mutagenesis Kit. Mouse
Apoer2 cDNA (a gift from J. Nimpf, Medical University of Vienna,
Austria)encoding the full-length
CAC38356) minus the alternatively spliced 59 amino acids exon
19 was cloned in frame with GFP or HA as described above for
VLDLR. The FLAG-ApoER2(WT) construct was generated by
subcloning ApoER2-HA into pCMV-Tag (Stratagene). This
construct was further mutagenized to generate the FLAG-
ApoER2(R774L) in which Arg residue 774 is replaced by a Leu.
All constructs were sequenced to verify the intended mutations.
Mouse Dab1 cDNA (accession # NP_796233) encoding the 555
amino acids isoform 2 was HA-tagged by PCR and subcloned into
pcDNA3.1 vector (Invitrogen). Mouse Pafah1b3 cDNA (accession
#Q61205) encoding the 29 kDa a1 subunit, mouse Pafah1b2
cDNA (accession # Q61206) encoding the 30 kDa a2 subunit, and
mouse Pafah1b1 cDNA (accession #NP_038653) encoding the 45
kDa b1 subunit of the Pafah1b complex, were cloned into the
pEGFP-C1 (Clontech), pcDNA3.1(+)-myc/his (Invitrogen) or
pCMV-Tag (Stratagene) to introduce the GFP, Myc or FLAG
Cell culture and protein analysis
COS7 or 293T cells (ATCC) were cultured in DMEM supple-
mented with 10% fetal bovine serum, and transfected with
expression vectors using the Fugene 6 reagent (Roche). After 30–
40 hours, the cells were harvested and the proteins extracted in
lysis buffer (PBS, 5 mM EDTA, 1% Triton X-100, pH 7.4) in the
presence of protease inhibitors (Mini Complete protease inhibitor
cocktail tablets, Roche). For immunoprecipitation, the lysates were
incubated with appropriate antibodies for 1–2 hours at 4uC,
followed by protein A/G agarose beads (Pierce). Samples were
analyzed by SDS-PAGE. To assay Reelin signaling, primary
Figure 6. reeler-like disruption of hippocampal layers in Pafah1b1+/2;
Apoer22/2double mutants. Comparable sagittal sections of the
hippocampus obtained from adult mice of the indicated genotype
were stained with cresyl violet. The hippocampus proper (HP) and the
dentate gyrus (DG) of Apoer2+/2mice are normal, whereas a splitting of
the pyramidal layers is evident in Pafah1b1+/2and in Apoer22/2mice.
Severe dyslamination of cellular layers is seen in double Pafah1b1+/2;
Apoer22/2(reeler-like). Scale bar=500 mm.
Figure 7. Reelin induces Dab1 and Akt phosphorylation in Pafah1b1+/2;
Apoer22/2double mutant neurons. Cortical neurons were cultured
from mutant mice of the indicated genotype, and incubated with either
control or Reelin-containing medium for 20 min. Lysates were analyzed
by Western blot using the 4G10 antibody to detect Dab1 phosphor-
ylation on tyrosine residues and a phospho-Akt antibody to detect Akt
phosphorylation on serine 473. Blots were reprobed with antibodies
against total Dab1 and Akt to ensure that similar amount of proteins
were present in each sample, and with antibodies against VLDLR and
ApoER2 to confirm the genotype of the mutants.
Pafah1b Binds Vldlr
PLoS ONE | www.plosone.org7February 2007 | Issue 2 | e252
cortical neurons were cultured from embryonic mice and treated
with Reelin-containing conditioned medium for 20 min. Cells
were lysed and proteins were subjected to Western blot analysis as
previously described .
In vitro binding assay
Vldlr, Apoer2, Pafah1b1, Pafah1b3, Pafah1b2, and Dab1 cDNAs were
produced in vitro using the TNT Quick Coupled Transcription/
Translation System (Promega), according to the manufacturer
instructions using35S-labeled methionine (Amersham Biosciences).
Proteins were separated by SDS-PAGE and detected by autoradi-
ography on dried gels. Quantitative analysis of autoradiography
bands density was performed using ImageJ software (NIH image).
Reeler mutant mice were obtained from The Jackson Laboratories
on a C57BL/66C3H hybrid background. Vldlr, Apoer2 and Dab1
knock out mice were on a hybrid C57BL/66129S6/SvEv. Pafah1b1
SvEv background. Mutants were genotyped by PCR as described
previously for Pafah1b1 , Reelin , Apoer2 and Vldlr , and
Quantification and statistical analysis of cortical
Sagittal paraffin sections (5 mm) of the brain from siblings (when
possible) of each genotype were stained with cresyl violet or
processed for immunohistochemistry using antibodies against
Figure 8. Integrated model of Reelin and Lis1 signaling. Reelin binds to VLDLR and ApoER2 and causes src-family kinase (SFK) activation and Dab1
phosphorylation. Dab1 binds to the NPxY motif of both, VLDLR and ApoER2. Upon Reelin stimulation, phosphoDab1 (P-Dab1) interacts with Lis1 and
with other signal transduction molecules (grey circles). Lis1 also binds the catalytic subunits of the Pafah1b complex (a1 and a2) as well as
components of the cytoplasmic dynein complex (yellow square). a1 and a2 also bind VLDLR at the NPxYL motif and compete with Dab1 for receptor
occupancy. These subunits do not recognize the NPxYR motif of ApoER2. A unique domain of ApoER2 enables unique interactions with synaptic and
trafficking proteins (white octagons). The binding of catalytic Pafah1b subunits to VLDLR may displace P-Dab1 and promote its interaction with Lis1.
Signaling molecules downstream of Lis1 and Dab1 affect cytoskeleton dynamics by acting on microtubules (thick lines) or actin filaments (thin lines),
thereby controlling neuronal migration and layer formation.
Pafah1b Binds Vldlr
PLoS ONE | www.plosone.org8February 2007 | Issue 2 | e252
Calbindin (layer II–III) or Foxp2 (layer VI). Four paramedian
sagittal images at a level caudal to the corpus callosum were
utilized for quantitative analysis as previously described .
We thank T. Curran and J. Nimpf for plasmid constructs, G. Eichele for
Pafah1b3 mutant mice, B.A. Antalffy for help with histology.
Conceived and designed the experiments: GC GD. Performed the
experiments: GZ AA UB GZ. Analyzed the data: RM. Contributed
reagents/materials/analysis tools: JH AW AW. Wrote the paper: GD.
1. Reiner O, Carrozzo R, Shen Y, Wehnert M, Faustinella F, et al. (1993) Isolation
of a Miller-Dieker lissencephaly gene containing G protein beta-subunit-like
repeats. Nature 364: 717–721.
2. Hong SE, Shugart YY, Huang DT, Al Shahwan S, Grant PE, et al. (2000)
Autosomal recessive lissencephaly with cerebellar hypoplasia is associated with
human RELN mutations. Nature Genet 26: 93–96.
3. Rice DS, Curran T (2001) Role of the Reelin signaling pathway in central
nervous system development. Ann Rev Neurosci 24: 1005–1039.
4. Tissir F, Goffinet AM (2003) Reelin and brain development. Nat Rev Neurosci
5. D’Arcangelo G (2005) The reeler mouse: anatomy of a mutant. In:
Dhossche DM, ed. International Review of Neurobiology. San Diego, CA:
Elsevier Inc. pp. 383–417.
6. Howell BW, Hawkes R, Soriano P, Cooper JA (1997) Neuronal position in the
developing brain is regulated by mouse disabled-1. Nature 389: 733–736.
7. Sheldon M, Rice DS, D’Arcangelo G, Yoneshima H, Nakajima K, et al. (1997)
Scrambler and yotari disrupt the disabled gene and produce a reeler-like phenotype in
mice. Nature 389: 730–733.
8. Ware ML, Fox JW, Gonzales JL, Davis NM, Lambert de Rouvroit C, et al.
(1997) Aberrant splicing of a mouse disabled homolog, mdab1, in the scrambler
mouse. Neuron 19: 239–249.
9. Trommsdorff M, Gotthardt M, Hiesberger T, Shelton J, Stockinger W, et al.
(1999) Reeler/Disabled-like disruption of neuronal migration in knockout mice
lacking the VLDL receptor and ApoE receptor 2. Cell 97: 689–701.
10. Kuo G, Arnaud L, Kronstad-O’Brien P, Cooper JA (2005) Absence of Fyn and
Src causes a reeler-like phenotype. J Neurosci 25: 8578–8586.
11. Hirotsune S, Fleck MW, Gambello MJ, Bix GJ, Chen A, et al. (1998) Graded
reduction of Pafah1b1 (Lis1) activity results in neuronal migration defects and
early embryonic lethality. Nat Genet 19: 333–339.
12. Gambello MJ, Darling DL, Yingling J, Tanaka T, Gleeson JG, et al. (2003)
Multiple dose-dependent effects of Lis1 on cerebral cortical development.
J Neurosci 23: 1719–1729.
13. Hattori M, Adachi H, Tsujimoto M, Arai N, Inoue K (1994) Miller-Dieker
lissencephaly gene encodes a subunit of brain platelet-activating factor
acetylhydrolase. Nature 370: 216–218.
14. Albrecht U, Abu-Issa R, Ratz B, Hattori M, Aoki J, et al. (1996) Platelet-
activating factor acetylhydrolase expression and activity suggest a link between
neuronal migration and platelet-activating factor. Dev Biol 180: 579–593.
15. Ho YS, Swenson L, Derewenda U, Serre L, Wei Y, et al. (1997) Brain
acetylhydrolase that inactivates platelet-activating factor is a G-protein-like
trimer. Nature 385: 89–93.
16. Yan W, Assadi AH, Wynshaw-Boris A, Eichele G, Matzuk MM, et al. (2003)
Previously uncharacterized roles of platelet-activating factor acetylhydrolase 1b
complex in mouse spermatogenesis. Proc Natl Acad Sci U S A 100: 7189–7194.
17. Koizumi H, Yamaguchi N, Hattori M, Ishikawa TO, Aoki J, et al. (2003)
Targeted disruption of intracellular type I platelet activating factor-acetylhy-
drolase catalytic subunits causes severe impairment in spermatogenesis. J Biol
Chem 278: 12489–12494.
18. Nothwang HG, Kim HG, Aoki J, Geisterfer M, Kubart S, et al. (2001)
Functional hemizygosity of PAFAH1B3 due to a PAFAH1B3-CLK2 fusion gene
in a female with mental retardation, ataxia and atrophy of the brain. Hum Mol
Genet 10: 797–806.
19. Wynshaw-Boris A, Gambello MJ (2001) LIS1 and dynein motor function in
neuronal migration and development. Genes Dev 15: 639–651.
20. Faulkner NE, Dujardin DL, Tai CY, Vaughan KT, O’Connell CB, et al. (2000)
A role for the lissencephaly gene LIS1 in mitosis and cytoplasmic dynein
function. Nat Cell Biol 2: 784–791.
21. Smith DS, Niethammer M, Ayala R, Zhou Y, Gambello MJ, et al. (2000)
Regulation of cytoplasmic dynein behaviour and microtubule organization by
mammalian Lis1. Nat Cell Biol 2: 767–775.
22. Tsai JW, Chen Y, Kriegstein AR, Vallee RB (2005) LIS1 RNA interference
blocks neural stem cell division, morphogenesis, and motility at multiple stages.
J Cell Biol 170: 935–945.
23. Assadi AH, Zhang G, Beffert U, McNeil RS, Renfro AL, et al. (2003) Interaction
of reelin signaling and Lis1 in brain development. Nat Genet 35: 270–276.
24. Andersen OM, Yeung CH, Vorum H, Wellner M, Andreassen TK, et al. (2003)
Essential role of the apolipoprotein E receptor-2 in sperm development. J Biol
Chem 278: 23989–23995.
25. Benhayon D, Magdaleno S, Curran T (2003) Binding of purified Reelin to
ApoER2 and VLDLR mediates tyrosine phosphorylation of Disabled-1. Mol
Brain Res 112: 33–45.
26. Howell BW, Herrick TM, Cooper JA (1999) Reelin-induced tyrosine
phosphorylation of Disabled 1 during neuronal positioning. Genes Dev 13:
27. Keshvara L, Benhayon D, Magdaleno S, Curran T (2001) Identification of
reelin-induced sites of tyrosyl phosphorylation on disabled 1. J Biol Chem 276:
28. Ballif BA, Arnaud L, Arthur WT, Guris D, Imamoto A, et al. (2004) Activation
of a Dab1/CrkL/C3G/Rap1 pathway in Reelin-stimulated neurons. Curr Biol
29. Beffert U, Morfini G, Bock HH, Reyna H, Brady ST, et al. (2002) Reelin-
mediated signaling locally regulates protein kinase B/Akt and glycogen synthase
kinase 3beta. J Biol Chem 277: 49958–49964.
30. Bock HH, Jossin Y, Liu P, Forster E, May P, et al. (2003) PI3-Kinase interacts
with the adaptor protein Dab1 in response to Reelin signaling and is required for
normal cortical lamination. J Biol Chem 278: 38772–38779.
31. Ballif BA, Arnaud L, Cooper JA (2003) Tyrosine phosphorylation of Disabled-1
is essential for Reelin-stimulated activation of Akt and Src family kinases. Brain
Res Mol Brain Res 117: 152–159.
32. Pramatarova A, Ochalski PG, Chen K, Gropman A, Myers S, et al. (2003) Nck
beta interacts with tyrosine-phosphorylated disabled 1 and redistributes in
Reelin-stimulated neurons. Mol Cell Biol 23: 7210–7221.
33. Huang Y, Magdaleno S, Hopkins R, Slaughter C, Curran T, et al. (2004)
Tyrosine phosphorylated Disabled 1 recruits Crk family adapter proteins.
Biochem Biophys Res Commun 318: 204–212.
34. Suetsugu S, Tezuka T, Morimura T, Hattori M, Mikoshiba K, et al. (2004)
Regulation of actin cytoskeleton by mDab1 through N-WASP and ubiquitina-
tion of mDab1. Biochem J 384: 1–8.
35. D’Arcangelo G, Homayouni R, Keshvara L, Rice DS, Sheldon M, et al. (1999)
Reelin is a ligand for lipoprotein receptors. Neuron 24: 471–479.
36. Hiesberger T, Trommsdorff M, Howell BW, Goffinet AM, Mumby MC, et al.
(1999) Direct binding of Reelin to VLDL receptor and ApoE receptor 2 induces
tyrosine phosphorylation of Disabled-1 and modulates Tau phosphorylation.
Neuron 24: 481–489.
37. Stockinger W, Brandes C, Fasching D, Hermann M, Gotthardt M, et al. (2000)
The reelin receptor ApoER2 recruits JNK-interacting proteins-1 and -2. J Biol
Chem 275: 25625–25632.
38. Gotthardt M, Trommsdorff M, Nevitt MF, Shelton J, Richardson JA, et al.
(2000) Interactions of the low density lipoprotein receptor gene family with
cytosolic adaptor and scaffold proteins suggest diverse biological functions in
cellular communication and signal transduction. J Biol Chem 275:
39. Brandes C, Kahr L, Stockinger W, Hiesberger T, Schneider WJ, et al. (2001)
Alternative splicing in the ligand binding domain of mouse ApoE receptor-2
produces receptor variants binding reelin but not alpha 2-macroglobulin. J Biol
Chem 276: 22160–22169.
40. Beffert U, Weeber EJ, Durudas A, Qiu S, Masiulis I, et al. (2005) Modulation of
synaptic plasticity and memory by Reelin involves differential splicing of the
lipoprotein receptor ApoER2. Neuron 47: 567–579.
41. Boycott KM, Flavelle S, Bureau A, Glass HC, Fujiwara TM, et al. (2005)
Homozygous deletion of the very low density lipoprotein receptor gene causes
autosomal recessive cerebellar hypoplasia with cerebral gyral simplification.
Am J Hum Genet 77: 477–483.
42. Niu S, Renfro A, Quattrocchi CC, Sheldon M, D’Arcangelo G (2004) Reelin
promotes hippocampal dendrite development through the VLDLR/ApoER2-
Dab1 pathway. Neuron 41: 71–84.
Pafah1b Binds Vldlr
PLoS ONE | www.plosone.org9 February 2007 | Issue 2 | e252