Articles |
From the Centro Prevenzione Arteriosclerosi, Dipartimento di Medicina Interna (S.B., M.R., N.E.), and the Istituto di Biologia e Genetica (D.C., C.C.), Università di Genova; the Sezione di Patologia Generale, Dipartimento di Scienze Biomediche, Università di Modena (R.G., W.L., R.M.T., M.G., M.L.S., P.M., S.C.); the Istituto di Terapia Medica Sistematica, Università di Roma (C.S.); and Ospedale S. Bernardino, Passignano sul Trasimeno (G.O.), Perugia, Italy.
Correspondence to Dr S. Calandra, Sezione di Patologia Generale, Dipartimento di Scienze Biomediche, Università di Modena, Via Campi 287, 41100 Modena, Italy.
| Abstract |
|---|
|
|
|---|
Key Words: Southern analysis polymerase chain reaction deletions LDL-receptor mRNA direct sequencing genetic screening
| Introduction |
|---|
|
|
|---|
| Methods |
|---|
|
|
|---|
|
Southern and Northern Blot Analyses
Southern blot analysis of DNA extracted from blood
leukocytes was carried out as previously reported.8 10
Filters were hybridized with LDL-R cDNA probes as previously
described.8
Total cellular RNA was extracted in guanidine-thiocyanate11 from cultured skin fibroblasts that had been maintained in lipoprotein-deficient serum for 15 hours.9 Northern blot analysis was performed as previously reported.8
Reverse Transcription and PCR Amplification
Cellular RNA (1 µg) was reverse-transcribed in a 20-µL
reaction mixture containing 8 mmol/L MgCl2, 1 mmol/L
of each dNTP, 1 U RNAsin, 100 pmol random hexamers, and 12 U avian
myeloblastosis virus (AMV) reverse transcriptase in 1x PCR buffer (30
mmol/L KCl, 50 mmol/L Tris-HCl, pH 9).12 After the sample
was heated at 95°C for 5 minutes, 80 µL of 1x polymerase chain
reaction (PCR) buffer containing 20 pmol of each primer was added as
well as 2.6 U Taq DNA polymerase. MgCl2 concentration in
the reaction mixture ranged from 1.5 to 3.5 mmol/L. The following
primers were used: (1) 5' GCT GGA AAT TGC GCT GGA CCG TCG C 3' (forward
primer in exon 1); (2) 5' CAT GAA CAG GAT CCA CCA C 3' (reverse primer
in exon 10); (3) 5' ATG ACA CCG TCA TCA GCA GGG A 3' (forward primer in
exon 10); (4) 5' GTT GTG GAA GAG GAC CAT ATC 3' (reverse primer in exon
13); (5) 5' ACC GGA AGA CCA TCT TGG AGG ATG A 3' (forward primer in
exon 12); (6) 5' CTT GGT GAG ACA TTG TCA CTA TC 3' (reverse in primer
in exon 15); (7) 5' GAG TGA ACT GGT GTG AGA G 3' (forward primer in
exon 14); and (8) 5' CAA GGC CGG CGA GGT CTC AGG A 3' (reverse primer
in exon 18). The conditions were 95°C for 5 minutes, 55°C for 2
minutes, and 72°C for 3 minutes for the first time, and subsequently
94°C for 2 minutes, 55°C for 2 minutes, and 72°C for 3 minutes
for 28 cycles. PCR products were separated by electrophoresis through a
1.5% agarose gel and sequenced directly with the fmol Sequencing
System (Promega Co). Primers used in the sequencing reactions were exon
1 forward primer, exon 10 forward primer, exon 12 forward primer (as
specified above), and exon 17 reverse primer (5' TTC TTC CAT AGA AGG
AAG ACC 3').
PCR Amplification of Genomic DNA
To amplify the region of LDL-R gene encompassing the deletion
joint in FHRoma,
FHChieti/Macerata, and FHGenova,
1 µg of genomic DNA was amplified in a 100-µL mixture containing
0.2 mmol/L of each dNTP, 50 pmol of each primer, 2.5 U Taq
DNA polymerase in 1x PCR buffer (see above), 1.5 to 5 mmol/L
MgCl2. Exon 12 forward primer and exon 15 reverse primer
(see above) were used for FHRoma and
FHChieti/Macerata. Exon 10 forward primer (see above) and
intron 13 reverse primer (5' GTT TCC ACA AGG AGG TTT CAA GGTT 3') were
used for FHGenova. The conditions were 95°C for 5
minutes, 50°C for 90 seconds, 72°C for 120 seconds, and 95°C for
90 seconds for 30 cycles. PCR products were separated by 1% agarose
gel electrophoresis after ethidium bromide staining. For
FHRoma and FHChieti/Macerata, the
appropriate fragments were excised and DNA was extracted from the gel
with Quiaex (Diagen, GmbH). Approximately 100 ng of each fragment was
digested with 10 UI of XbaI and
ApaLI. The fragments obtained after digestion were
separated by 1% agarose gel and stained with ethidium bromide.
Ligand Blotting
Ligand blotting experiments were performed in proband P.F.
(FHPadova) by using protein extracted from cultured
fibroblasts and either human 125I-LDL or rabbit
125I-ß VLDL as ligands.13 14 Lipoprotein
labeling was carried out according to a standard
procedure.9 Membranes were subjected to autoradiography on
Hyperfilm-MP (Amersham, UK) for 2 to 4 hours. Densitometric
analysis of radiograph films was carried out with an LKB-XL laser
densitometer (LKB, Bromma). The radioactive bands corresponding to
LDL-R (130 kD) were excised from the membrane and counted directly in a
Beckman 5000 gamma counter.
| Results |
|---|
|
|
|---|
Deletion of Exons 3 Through 10 (FHMassa)
Southern blot analysis of proband G.F. (Table 1
) DNA digested
with several restriction enzymes and hybridized with cDNA probes
complementary to exons 1 through 4, 5 through 7, 8 through 10, and 11
through 14 suggested the presence of a large deletion (12 kb) that
spanned from the 3' end of intron 2 to the 5' end of intron 11 (Fig 1
). Northern blot analysis showed a reduced
intensity of the normal LDL-R mRNA (5.3 kb) but failed to reveal the
presence of mutant mRNA. RNA was reverse-transcribed and amplified by
using exon 1 forward primer and exon 15 reverse primer (see
"Methods"). Four fragments of approximately 2300, 900, 780, and
640 bp were detected in agarose gel (Fig 2
). The
nucleotide sequence showed that (1) the 2300-bp fragment corresponded
to the normal mRNA (exon 2-exon 3 junction) (Fig 3A
); (2) in the 900-bp
fragment, exon 2 was followed by exon 11, as expected from the Southern
blot results (Fig 3A
); (3) in the 640-bp fragment, exon
2 was followed by exon 13 (Fig 3A
). We could not obtain a sufficient
amount of the 780-bp fragment for sequence analysis. The size of
this fragment, however, suggests that it might derive from the joining
of exon 2 to exon 12. Thus, the deletion-bearing allele of proband GF
generates three mRNA species, the most abundant of which (900-bp
fragment) encodes a peptide that contains the 42 amino acids of the
NH2 terminal end of the LDL-R and 18 novel amino acids
(indicated in italics in Table 2
) preceding a stop
codon. The other abnormal mRNA species we sequenced (exon 2-exon 13) is
expected to encode a protein that contains only 42 amino acids (Table 2
). The joining of exon 2 to exon 12 (the 780-bp fragment that was not
sequenced) would produce a short receptor devoid of part of the ligand
and of EGF precursor homology domains (Table 2
). This new deletion
found in a single family living in the northwestern part of Tuscany
(central Italy) has been designated FHMassa.
|
|
|
|
Deletion of Exons 11 and 12 (FHGenova)
Southern blot analysis of proband B.M. (Table 1
) DNA digested
with several restriction enzymes and hybridized with exon 11 through 18
probe suggested the presence of a deletion of
4 kb spanning from the
5' half of intron 10 to the 3' end of intron 12 and eliminating exons
11 and 12 (Fig 1
). PCR amplification of proband genomic DNA obtained by
using exon 10 forward primer and intron 13 reverse primer generated a
fragment of 2800 bp (data not shown), the size of which was consistent
with a 4-kb deletion (Fig 1
). Northern blot analysis showed the
presence of a single band corresponding to the normal LDL-R mRNA (data
not shown). Proband RNA was reverse-transcribed and amplified by using
exon 10 forward primer and exon 13 reverse primer. Two fragments of 580
and 320 bp of comparable intensity (data not shown) were obtained and
sequenced directly. In the larger fragment, corresponding to the normal
mRNA, exon 10 continued with exon 11, whereas in the smaller fragment,
exon 10 was followed by exon 13 (Fig 3A
). The junction between exon 10
(GG) and exon 13 (GAC) generates a new codon (GGG) that encodes the
same amino acid (glycine) as the codon generated by the junction of
exon 10 with exon 11 (GGC). However, the junction between exon 10 and
exon 13 causes a shift in the reading frame, which produces a sequence
of 48 novel amino acids preceding a stop codon (Table 2
). This new
deletion was named FHGenova after the city of origin of the
proband's family.
Deletions of Exons 13 and 14 (FHRoma and
FHChieti/Macerata)
FHRoma
Southern blot analysis of proband M.S. (Table 1
) DNA suggested
the presence of a deletion of approximately 5 kb downstream of exon 12
(Fig 1
). To define the boundaries of the deletion more precisely, we
used the following strategies: (1) PCR amplification of genomic DNA
using exon 12 forward primer and exon 13 reverse primer, followed by
the digestion with ApaLI and Xba I; and
(2) PCR amplification of genomic DNA using exon 12 forward primer and
exon 15 reverse primer, followed by digestion with
ApaLI15 16 (Fig 4
).
Amplification of the exon 12exon 15 region, which was possible only
in proband DNA (in control DNA, this region is too large to be amenable
to PCR amplification), produced a fragment with a size (1800 bp)
compatible with a deletion of 4.7 kb (see Fig 4
for details). Northern
blot analysis of proband LDL-R mRNA showed the presence of the
normal mRNA band with approximately half the intensity of that found in
control cells (data not shown). Reverse transcription and amplification
of proband RNA with exon 12 forward primer and exon 15 reverse primer
produced only a normal fragment of
530 bp (data not shown). However,
when amplification was performed using exon 10 forward primer/exon 15
reverse primer, we could detect besides the normal fragment (900 bp) a
minute amount of an abnormal fragment (650 bp) (Fig 5
).
The sequence of the latter showed that exon 12 joined exon 15 (Fig 3B
),
as expected. This new deletion was named FHRoma after the
city of origin of proband's family.
|
|
FHChieti/Macerata
We previously reported this deletion in two members of a family
living in the central districts of Italy
(FHChieti)8 ; a group from Dallas, Tex,
reported this deletion in a homozygous Canadian patient of Italian
ancestry.17 The same mutation (see below) has recently
been found in three other unrelated subjects (M.L., G.F., M.M.) (Table 1
) living in central Italy. Although this deletion eliminates exons 13
and 14, its size and boundaries are different from those defined in
FHRoma (Fig 4
). More specifically, the PCR amplification of
the exon 12exon 15 region from genomic DNA, followed by digestion
with ApaLI, showed that in the
FHChieti/Macerata mutation, the deletion (3.7 kb) was
smaller than that of FHRoma (Fig 4
). As found in
FHRoma, reverse transcription of RNA and PCR
amplification of the exon 10exon 15 region produced a trace amount of
a 650-bp fragment and the normal 900-bp fragment (Fig 5
). The size of
the 650-bp fragment is consistent with the idea that the mutant mRNA in
FHChieti/Macerata lacks exons 13 and 14. This was confirmed
by the direct nucleotide sequence in Fig 3B
.
Analysis of the restriction fragment length polymorphisms (RFLPs) of
the LDL-R gene in the four patients found to be carriers of this
deletion (Table 1
) revealed that the haplotype cosegregating with the
mutant allele [Sph I (+), Stu I (+),
HincII (+), ApaLI 5' (+),
Pvu II (-), Nco I (+), ApaLI
3' (+)] was the same in all affected subjects. Comparison of this
haplotype with that found in the Italian-Canadian patient
[Bms I (+), Sph I (+), Stu (+),
ApaLI 5' (+), Pvu II (-), Nco
I (+), ApaLI 3' (+), H.H. Hobbs, personal
communication, 1994] strongly suggests that all probands with
FHChieti/Macerata deletion shared a common ancestor who had
presumably come from the central-eastern districts of Italy, close to
Chieti and Macerata.
The joining of exon 12 to exon 15 in mRNA encoded by either the
FHRoma or FHChieti/Macerata allele produces a
shift of the reading frame, leading to a sequence of 15 novel amino
acids before a premature stop codon (Table 2
). The amount of this
truncated receptor in proband cells is expected to be negligible in
view of the fact that in both FHRoma and
FHChieti/Macerata, the level of mutant LDL-R mRNA is
very low (Fig 5
).
Deletion of Exon 16 (FHPadova-2)
Southern blot analysis of proband DNA (P.F.) (Table 1
)
digested with several restriction enzymes and hybridized with various
exon-specific cDNA probes suggested the presence of a 2-kb deletion
spanning from the HindIII site in intron 15 to the
KpnI site in intron 16 and eliminating the sole exon 16 (Fig 1
). Northern blot analysis of proband RNA showed the presence of
the 5.3-kb band with an intensity comparable to that found in control
cells (data not shown). Reverse transcription of proband mRNA and
amplification of the exon 14exon 18 region produced comparable
amounts of two fragments of 600 and 520 bp (data not shown). The
nucleotide sequence showed that the 600-bp fragment corresponded to the
normal mRNA (exon 15exon 16 junction), whereas the 520-bp fragment
corresponded to the mutant mRNA (exon 15exon 17 junction) (Fig 3B
).
The joining of exon 15 with exon 17 does not change the reading frame
(Table 2
). Thus, the mutant mRNA is translated into a receptor protein
of 813 amino acids devoid of the last 18 amino acids downstream of the
O-linked sugar domain and the first 8 amino acids of the transmembrane
domain. Ligand blotting experiments using rabbit 125I-ß
VLDL (Fig 6
) and human 125I-LDL (data not
shown) indicated that proband fibroblasts bound half the amount of both
ligands compared with control cells. This new deletion has been
designated FHPadova-2 after the city of origin of the
proband's ancestors. A point mutation designated FHPadova
(Asp200
Gly) was reported previously by Hobbs et
al4 in an Italian-American FH patient. To avoid
misunderstanding, we renamed this mutation FHPadova-1.
|
| Discussion |
|---|
|
|
|---|
Gln at codon 3500).18 These deletions, all found in
heterozygous patients, were characterized by using a combination of
techniques such as Southern blotting, PCR amplification of genomic DNA
(whenever possible), Northern blot analysis, reverse transcription,
and PCR amplification of LDL-R mRNA (RT-PCR) extracted from skin
fibroblasts. In all cases, the use of RT-PCR allowed us to identify the
mutant mRNAs even in those situations in which this mRNA had not been
detected in Northern blot analysis because of either a small size
difference with respect to its normal counterpart or its reduced
content. From analysis of the RT-PCR products, three different
situations emerged: (1) in FHGenova (deletion of exons 11
and 12) and FHPadova-2 (deletion of exon 16), the amount of
the mutant mRNA was apparently identical to that of its normal
counterpart; (2) in FHRoma and
FHChieti/Macerata (deletion of exons 13 and 14), the mutant
mRNA was present in trace amounts; and (3) in FHMassa
(deletion of exons 2 through 10), the mutant mRNA consisted of one
major and two minor forms (in the major form, exon 1 joined exon 11, as
was expected from the genomic DNA analysis, whereas the size of the
two minor forms was consistent with the skipping of either exon 11 or
exon 12, caused by an alternative splicing). These findings lend
further support to the general idea that the intracellular level of
mutant LDL-R mRNA varies considerably in relation to both the type of
mutation and the occurrence of alternative
splicings.8 12
The sequence of the deletion joints shows that the deletion-bearing
alleles encode either a very short peptide (FHMassa), a
truncated receptor (FHGenova,
FHRoma, and FHChieti/Macerata), or a
receptor with an in-frame deletion of few amino acids
(FHPadova-2) (Table 2
). Because it is conceivable that
truncated receptors lacking several important domains are rapidly
degraded intracellularly, the functional defects caused by these
deletions belong to either class I or II.3 4 In
FHPadova-2 (deletion of the sole exon 16), the expected
mutant receptor protein is devoid of the last 18 amino acids downstream
the O-linked region (ie, those amino acids immediately outside the
plasma membrane) and the proximal 8 amino acids of the membrane
spanning domain. Because the results of the ligand blot with rabbit
125I ß-VLDL showed that in the proband fibroblasts there
was half the amount of the normal receptor, it is most likely that this
mutation reduces the amount of receptor protein available for ligand
binding. The membrane spanning domain of the mutant receptor would
theoretically contain 8 novel amino acids belonging either to the
immediately extracellular region of the O-linked sugar or to the
proximal region of the cytoplasmic domain. In this new configuration,
the average hydropathy index19 of the novel membrane
domain (1.93 to 1.51) would drop below the values usually found in the
plasma membrane domain of LDL-R observed in several animal
species,20 21 22 23 24 which ranges from 2.123 to
2.54.21 22 It is reasonable to assume that in
FHPadova-2 the receptor protein may not be inserted in the
plasma membrane, even though it is safely transported through the
cytoplasm.
The fact that the deletion of exon 16 is a cause of FH is supported by a previous observation by some Dutch investigators who reported a deletion of this exon in a heterozygous FH patient with a 0.4-kb deletion spanning from intron 15 to the 5'-splice site of exon 16.25
It has not been the aim of this study to analyze the molecular mechanisms responsible for the deletions reported here. It is reasonable to assume, however, that Alu sequence recombinations might be the cause of at least three of the five deletions described: FHRoma, FHChieti/Macerata, and FHPadova-2. Alu sequences are indeed present in introns 12 and 14 (FHChieti/Macerata and FHRoma) and in introns 15 and 16 (FHPadova-2).26 27 28
We have taken advantage of the description of the four new
deletions (FHMassa, FHGenova,
FHRoma, FHPadova-2) to update the series
of deletions found so far in Italian FH patients (Table 3
). In our series, large deletions account for
approximately 7% of all mutations of LDL-R gene, a value fairly
similar to that observed in other populations.5 6 During
our survey, we confirmed that some deletions (eg, the previously
reported FHChieti/Macerata and
FHPavia)8 29 were clustered in specific areas
of Italy (northwestern districts for FHPavia;
central-eastern districts for FHChieti/Macerata). Thus,
despite the genetic heterogeneity of the Italian population, clusters
of identical mutations of LDL-R gene are not uncommon.
|
The small number of patients carrying the same mutation (Table 3
) does not allow us to draw firm conclusions on the genotype-phenotype
relationship. However, it is of interest that among the probands
carrying the FHChieti/Macerata mutation, subject M.M. is
free of the typical clinical features of FH (tendon xanthoma, arcus
cornealis, and coronary heart disease) despite a fairly high level of
plasma LDL cholesterol. A possible explanation for this finding may be
the sex and the relatively low Lp(a) level of proband M.M. A survey of
the patients of our series confirmed that the clinical features of FH
are much less severe and delayed in female FH patients compared with
males. On the other hand, the level of Lp(a) in our patients shows a
strong correlation with coronary heart disease and carotid
atherosclerosis but not with tendon xanthomatosis or arcus cornealis
(S.B., M.R., S.C., unpublished observations).
As a corollary to our study, we propose the use of
nonradioactive procedures for the identification of some deletions in
Table 3
. The systematic screening of three deletions
(FHGenova, FHRoma, and
FHChieti/Macerata) may be successfully performed by PCR
amplification of genomic DNA, followed (if necessary) by the digestion
of the amplified fragment with appropriate restriction enzymes, as
illustrated in Fig 4
for FHChieti/Macerata and
FHRoma.
| Acknowledgments |
|---|
Received July 27, 1994; accepted October 31, 1994.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Bertolini, A. Cantafora, M. Averna, C. Cortese, C. Motti, S. Martini, G. Pes, A. Postiglione, C. Stefanutti, I. Blotta, et al. Clinical Expression of Familial Hypercholesterolemia in Clusters of Mutations of the LDL Receptor Gene That Cause a Receptor-Defective or Receptor-Negative Phenotype Arterioscler. Thromb. Vasc. Biol., September 1, 2000; 20 (9): e41 - e52. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. E Heath, I. N M Day, and S. E Humphries Universal primer quantitative fluorescent multiplex (UPQFM) PCR: a method to detect major and minor rearrangements of the low density lipoprotein receptor gene J. Med. Genet., April 1, 2000; 37(4): 272 - 280. [Abstract] [Full Text] |
||||
![]() |
S. H. Kim, J. H. Bae, J. J. Chae, U. K. Kim, S.-J. Choe, Y. Namkoong, H.-S. Kim, Y.-B. Park, and C. C. Lee Long-Distance PCR-based Screening for Large Rearrangements of the LDL Receptor Gene in Korean Patients with Familial Hypercholesterolemia Clin. Chem., September 1, 1999; 45(9): 1424 - 1430. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
ATVB Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |