The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360

Review
Molecular pathogenesis of subarachnoid haemorrhage
Baiping Zhang a , Kaare Fugleholm b , Lorna B. Day a , Shu Ye a ,
Roy O. Weller b , Ian N.M. Day a,∗
a Human Genetics Division, School of Medicine, Southampton University Hospital NHS Trust, Duthie Building (Mailpoint 808),
Tremona Road, Southampton SO16 6YD, UK
b Clinical Neurosciences Division, School of Medicine, Southampton University Hospital NHS Trust,

Tremona Road, Southampton SO16 6YD, UK

Received 5 November 2002; accepted 31 December 2002

Abstract
Subarachnoid haemorrhage (SAH) results from leakage of blood into the subarachnoid space and carries high morbidity and
mortality. However, there is limited understanding to date, of the risk factors, cellular, intermediate biochemical and genetic
traits predisposing to SAH. Nevertheless, in conjunction with improved methods of diagnostic imaging and less invasive
approaches to preventing aneurysmal rupture, there may be utility in gaining a better understanding of the pathogenesis and
in identifying pre-disease markers. Additionally, it is not impossible that drugs of value (e.g. matrix or endothelial modifiers)
could become available. Several different clinical subtypes can be recognised, distinguished by arterial or venous involvement,
presence of unruptured arterial aneurysms, and apparently ‘sporadic’ and ‘familial’ occurrences. Epidemiological risk factors
include alcohol consumption and smoking: hypertension is a risk factor for rupture. About 10% seem to reflect strong family
history and this subset may be particularly illuminating with respect to the molecular pathogenesis. Haemodynamic stress
and poor vascular structure may be the main mechanisms of pathogenesis. The epidemiological and statistical evidence
for familial megaphenic genes and modifier genes is reviewed. This review focuses on the pathogenesis, as opposed to
inflammatory response to SAH. It sets in context the roles of specific genes and their protein products, such as polycystin
(PKD1), fibrillin (FBN1), collagen III (COL3A1), elastin (ELN), collagen IV, protease inhibitor or ␣1-antitrypsin (PI) and
proteases. These considerations illustrate the shortfalls in current knowledge, the needs of future biochemical and cellular
research and their potential implications for future prevention of this often fatal condition.
© 2003 Elsevier Science Ltd. All rights reserved.
Index terms: Subarachnoid haemorrhage; Stroke; Collagen; Elastin; Aneurysm; Intracranial aneurysm

Contents
1. Clinical overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1342
1.1. General risk factors, causes and outcomes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1343
1.2. Evidence for familiality in subarachnoid haemorrhage and intracranial aneurysm . . . . . . . . . 1344
2. Tissue and cellular pathophysiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1345
2.1. Pathology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1345
2.2. Pathology of saccular aneurysms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1346

∗Corresponding author. Tel.: +44-23-8079-5063; fax: +44-23-8079-4264.

E-mail address: inmd@soton.ac.uk (I.N.M. Day).

1357-2725/03/$ – see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S1357-2725(03)00043-8
1342 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360

2.3. Pathogenesis and the stages involved in the formation of saccular aneurysms . . . . . . . . . . . . 1347
2.4. Cellular and molecular architecture of the vessel wall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1349
3. Molecular pathology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1350
3.1. PKD1 gene and polycystin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1350
3.2. COL3A1 and type III collagen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1352
3.3. Fibrillin-1 and other linkages in aortic and mixed vessel aneurysmal disease . . . . . . . . . . . . . 1352
3.4. Genomewide-linkage and haplotype association at chromosome 7q11 in the elastin gene . . 1353
3.5. Collagen IV and other basement membrane constituents. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1354
3.6. ␣1-Antitrypsin (protease inhibitor) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1355
3.7. Association studies in sporadic intracranial aneurysm and subarachnoid haemorrhage . . . . . 1355
4. Future lines of investigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1356
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1356
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1357

1. Clinical overview is identified, rebleed is prevented by exclusion of the

aneurysm from the circulation. The results of a re-
Subarachnoid haemorrhage (SAH) is a condition cent multicentre study show that the best outcome is
caused by the escape of blood from a cerebral artery achieved by occluding the aneurysm with coils via the
into the subarachnoid space along the surface of the endovascular route. The alternative method is open mi-
brain. Spontaneous SAH usually presents with a sud- crosurgery where a clip is placed across the aneurysm
den onset explosive headache followed by various neck. Patients almost invariably require neurointen-
degrees of deterioration of conscious level and of sive care treatment to counteract the effects of the sub-
neurological status. There are often warning symp- arachnoid blood (for example cerebral artery spasm,
toms caused by small leaks of blood (warning leak, hydrocephalus and brain swelling) and maintain ade-
sentinel bleed), but unfortunately these are most often quate blood supply to the brain.
recognised retrospectively. SAH has a multitude of Counselling is essential and based on the current
aetiologies but it is generally accepted that a ruptured understanding of the pathogenesis of SAH. On ac-
intracranial aneurysm is the source of spontaneous count of the severity of SAH, patients and/or relatives
SAH in about 80% of cases (Kalimo, Kaste, & Haltia, usually seek information regarding risk factors and
2002). The other causes and general and molecular heritability. With the poor understanding of the patho-
pathology are considered in more detail in the next genesis the neurosurgical advice has been based on
sections. SAH is diagnosed by a detailed history, sup- the known risk factors and known genetically deter-
ported by a CT scan of the head. Small CT-negative mined risk conditions (Fig. 1). The aim has been to
bleeds are diagnosed by lumbar puncture and CSF identify high-risk relatives in order to undertake ef-
spectroscopy to identify blood breakdown products, fective screening with the use of MRI or conventional
which can be detected between 12 h and 2 weeks after angiography. The current practice in most centres is to
the bleed (Vermeulen & van Gijn, 1990). The mortal- offer screening to first degree relatives in families with
ity of diagnosed cases is about 40% within the first more than one member with intracranial aneurysms,
30 days despite the best current therapy. This does and in addition to other relatives with risk conditions
not take the patients who die from undiagnosed SAH or a heavy risk factor load. The decision is often
into account. Less than 25% have a good functional difficult and screening is sometimes performed on a
outcome and the disease has dramatic impact on the psychological rather than rational indication. The de-
lives of carers and relatives (Pritchard, Foulkes, Lang, cision to offer screening has had to be been balanced
& Neil-Dwyer, 2001). Prevention of SAH is therefore with the significant mortality/morbidity of treatment
of paramount importance. The two main aims of cur- of unruptured aneurysms (The International Study
rent treatment are to prevent rebleed and to counteract of Unruptured Intracranial Aneurysms Investigators,
the effect of the initial bleed. Where an aneurysm 1998). As a result of the International Subarachnoid
 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
Fig. 1. Schematic overview of contributory factors to subarachnoid
haemorrhage.
Aneurysm Trial (ISAT) study (ISAT, 2002), the less
invasive intravascular treatment with coiling has been
found to be safer than open surgery and has thereby
potentially moved the balance towards screening more
relatives, although there are really no new methods
to identify relatives at risk. Identification of genes
Fig. 2. Diagram depicting the commonest sites of saccular aneurysms related to the circle of Willis.
1343
or biochemical, cellular or other markers involved
in the molecular pathogenesis of SAH would be of
fundamental importance, not only to identify relatives
at high risk, but also with a potential for screening
of individuals in the general population, who carry a
heavy load of external risk factors (Fig. 1).
1.1. General risk factors, causes and outcomes
Subarachnoid haemorrhage (SAH) accounts for
10% of cerebrovascular disease and has an annual
incidence of 10–15 per 100,000, with peak incidence
of non-traumatic SAH from 40 to 60 years of age and
an overall female/male ratio of 3/2 particularly after
the age of 50 years. The prevalence of unruptured
intracranial aneurysms detected at autopsy ranges
from 0.2 to 8.9% (Tomasello, D’Avella, Salpietro,
& Longo, 1998). This suggests that most intracranial
aneurysms do not rupture. The commonest sites for
intracranial aneurysms are shown on Fig. 2. Risk
factors include hypertension, cigarette smoking, low
body mass index, infection, excess consumption of
alcohol or coffee, cocaine and amphetamine abuse,
family history of intracranial aneurysms and some
1344 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
hereditary conditions (see further) (Isaksen, Egge,
Waterloo, Romner, & Ingebrigtsen, 2002; Kissela
et al., 2002).
Although it may be possible to determine the cause
of a subarachnoid haemorrhage at post mortem, a truer
picture of the causes of subarachnoid haemorrhage is
obtained by cerebral angiography. Defining the major
causes of subarachnoid haemorrhage by cerebral an-
giography reveals three categories of vascular patholo-
gies (Kalimo et al., 2002):
(a) Ruptured saccular aneurysm of a cerebral artery
in ∼80% of cases.
(b) Bleeding from an arteriovenous malformation in
5–10% of patients. Arteriovenous malformations
are usually present from birth and arise in a va-
riety of different forms. Neural crest cells are
major contributors to mesenchymal structures in
the head and neck including arteries (Etchevers,
Vincent, Le Douarin, & Couly, 2001) and disor-
ders of neural crest development may be a major
factor in the formation of arteriovenous malfor-
mations (Bhattacharya et al., 2001).
(c) No cause is identified in 10–15% (Fogelholm,
Hernesniemi, & Vapalahti, 1993) and in 2/3 of
these cases the bleed fulfils the criteria of a per-
imesencephalic SAH, which has a very low risk
of rebleed and a better outcome than aneurysmal
SAH (Marquardt, Niebauer, Schick, & Lorenz,
2000).
Other causes of subarachnoid haemorrhage include
rupture of a mycotic aneurysm in which the vessel
wall is weakened by bacterial (Frazee, Cahan, &
Winter, 1980) or fungal (Horten, Abbott, & Porro,
1976) infection and the artery ruptures, and trauma
in which a blood vessel in the subarachnoid space is
injured and bursts. SAH is also associated with aor-
tic coarctation and hypertension and with tumours,
vasculitis and bleeding diatheses. Trauma is probably
the most frequent cause of SAH, but the significance
of this type of SAH is difficult to assess as it usually
overshadowed by the effect of other simultaneous
traumatic brain lesions.
Of the above causes of SAH, the familial causes
may be the most illuminating to pathogenetic path-
ways because they may represent the paradigm of
single isolated gene lesions representing one protein
product in one pathway or structure important to the
maintenance of normal structure and function of the
cerebral vasculature.
1.2. Evidence for familiality in subarachnoid
haemorrhage and intracranial aneurysm
Some early (1950s–1970s) case reports suggested
familial occurrences of SAH and occurrences in
identical twins. The advent of computer tomogra-
phy scanning, then magnetic resonance imaging, led
to the pre-symptomatic recognition of intracranial
aneurysms and as a consequence, enhanced recogni-
tion of an apparently familial subset of intracranial
aneurysm (IA) and subarachnoid haemorrhage (SAH).
Approaching 300 families have been described in
the literature (e.g. De Braekeleer, Perusse, Cantin,
Bouchard, & Mathieu, 1996; Schievink, Schaid,
Rogers, Piepgras, & Michels, 1994).
Familial occurrence is important to our understand-
ing of the disease because if specific gene lesions
occur, their identification and characterisation of the
cognate protein product will define a pathway of
aetiology. Different molecular genetic approaches
may be suitable, dependent on whether heritability is
monogenic, oligogenic or polygenic. It is therefore
important to be certain both that there is heritabil-
ity and to establish what types of heritability occur.
Several different types of study have attempted to
clarify this important question. These have included
prevalence studies, cohort studies including incidence
studies of subarachnoid aneurysm or haemorrhage
in relatives, prognostic studies of familial versus
non-familial SAH, case–control and descriptive case
series analyses. Specific genetic studies have exam-
ined SAH occurrence in first and second degree rela-
tives. Characteristics of sporadic and familial disease
have also been compared. The principal conclusions
which have been drawn from these studies are as
follows. The case series have shown that frequency
of subarachnoid haemorrhage varies amongst racial
groups (Ronkainen et al., 1998). Prevalence of in-
tracranial aneurysms range from 4% (Raaymakers,
1999) to 10% (Kojima et al., 1998) in the familial
groups. The Finnish and Japanese studies both cal-
culated population prevalence that was lower than in
the familial groups (Kojima et al., 1998; Ronkainen
et al., 1998). However, in the Japanese study the pop-
ulation was highly selected, representing individuals
 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
presenting for evaluation for brain diseases, which
may account for the population prevalence observed
to be higher than in the familial groups of some of the
other studies. Clearly some of the frequency variation
has depended on the mechanisms of ascertainment,
and on the sensitivity and specificity of screening
methods used. For intracranial aneurysm, magnetic
resonance angiography shows 98% agreement with
conventional angiography (Ronkainen et al., 1995).
In two studies of incidence of SAH in relatives (Gaist
et al., 2000; Wang, Longstreth, & Koepsell, 1995)
all SAH patients of interest were ascertained from a
nationwide or specified geographical area and SAH
was confirmed using specific diagnostic criteria. The
nationwide study of SAH in relatives constructed
prospectively in Denmark (Gaist et al., 2000) showed
a three–five-fold increased incidence of SAH in first
degree relatives compared with the general popu-
lation. The retrospective study by Schievink et al.
(1995) showed a four-fold increase in risk for first
degree relatives, indicating that familial aggregation
of ruptured aneurysms is not fortuitous but represents
a distinctive epidemiological pattern. By contrast,
the only case–control study of patients and relatives
with SAH yielded an odds ratio of 1.8 for first de-
gree relatives. However, this study used self-reported
history rather than specific diagnostic criteria (Wang
et al., 1995). Few studies have attempted to deter-
mine pattern of inheritance. However, Schievink et al.
(1994) undertook a segregation analysis of published
pedigrees. This segregation analysis was unable to
distinguish between the various transmission models
but concluded that risk of subarachnoid haemorrhage
is genetically heterogeneous with some families be-
ing consistent with an autosomal dominant model of
transmission, some autosomal recessive and perhaps
some X-linked families. However, the authors also
recognised a number of possible biases which could
account for the failure of a single segregation model
to explain the heritable risk of subarachnoid haem-
orrhage, so the failure of fit to a single segregation
model cannot per se be taken as proof of more than
one mode of inheritance. A large population based ge-
nealogy study from the Saguenay Lac-Saint-Jean re-
gion permitting recognition of consanguinity found no
evidence suggesting autosomal recessive inheritance,
which consanguinity predisposes (De Braekeleer
et al., 1996). The observed coefficient of inbreeding
1345
was not higher in intracranial families than in a con-
trol population, but there was a diminishing frequency
of aneurysms between first, second and third degree
relatives supporting an autosomal dominant model.
It has also been recognised in these studies that
there are characteristic clinical and pathological fea-
tures of familial cases which seem to distinguish them
from sporadic cases. Specifically, familial aneurysms
rupture at a younger age, on an average 5 years earlier
(Bromberg, Rinkel, Algra, Limburg, & van Gijn, 1995;
Kasuya, Onda, Takeshita, Hori, & Takakura, 2000;
Leblanc, Melanson, Tampieri, & Guttmann, 1995;
Lozano & Leblanc, 1987; Ronkainen, Hernesniemi,
& Tromp, 1995; Schievink et al., 1995). Additionally,
the aneurysms in clearly familial cases rupture at a
smaller size (Lozano & Leblanc, 1987; Ronkainen,
Hernesniemi, et al., 1995). Lozano and Leblanc (1987)
showed using multivariate methods that 70% of fa-
milial intracranial aneurysms had ruptured by age 50
years compared with 43% of non-familial aneurysms.
Also, consistently, aneurysms of the anterior com-
municating artery complex are underrepresented and
those of the middle cerebral artery are overrepre-
sented among familial cases (Bromberg et al., 1995;
Kasuya et al., 2000; Leblanc et al., 1995; Lozano
& Leblanc, 1987; Ronkainen, Hernesniemi, et al.,
1995; Schievink et al., 1995). However, the studies
disagree about the evidence for female preponderance
and aneurysm multiplicity in familial intracranial
aneurysms, although aneurysm multiplicity does seem
to be a frequent observation in familial cases.
Similarly there is no agreement in prognostic studies
concerning the outcome in familial versus non-familial
subarachnoid haemorrhage using graded outcomes at
discharge from hospital or at 12 months after the bleed.
Nevertheless, these pathological and anatomical pat-
terns point to distinct and specific cellular and molecu-
lar pathology in families showing evidence of a single
megaphenic gene effect.
2. Tissue and cellular pathophysiology
2.1. Pathology
The majority of subarachnoid haemorrhages are due
to the rupture of arteries and especially rupture of sac-
cular aneurysms. In humans, the subarachnoid space
1346 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
contains cerebrospinal fluid and forms a water-jacket
around the brain. Large and medium sized arteries
and veins reside in the subarachnoid space before they
penetrate the surface of the brain to supply the cere-
bral cortex and deep white and grey matter. When
subarachnoid haemorrhage occurs, therefore, blood is
released into the subarachnoid space at arterial pres-
sure and spreads diffusely over the surface of the
brain bathing arteries in the subarachnoid space with
fresh blood (Kalimo et al., 2002). Under these cir-
cumstances arterial spasm occurs and this results in
cerebral ischaemia and infarction (Vermeulen & van
Gijn, 1990). Mortality rates for rupture of saccular
aneurysms are high either due to cerebral infarction or
aneurysms bursting into the brain causing intracerebral
haemorrhage. Re-bleeding from saccular aneurysms
1–7 days after the initial haemorrhage is common
thus increasing the mortality rate (Rosenorn, Eskesen,
Schmidt, & Ronde, 1987).
2.2. Pathology of saccular aneurysms
By definition, a saccular aneurysm is a berry-shaped
or multi-lobed outpouching on a major cerebral
artery usually associated with the circle of Willis
at the base of the brain. Such aneurysms may be
discovered incidentally at post mortem or may be
revealed as a result of subarachnoid haemorrhage.
They are distinguished by their shape and pathogene-
sis from fusiform aneurysms (due to dilatation of an
atherosclerotic artery especially the basilar artery);
mycotic aneurysms (due to focal necrosis of an artery
wall following bacterial or fungal infection; dissect-
ing aneurysms in which the blood ruptures into the
wall of an artery and separates its layers often result-
ing in occlusion of the vessel. Dissecting aneurysmas
of the carotid, middle cerebral and vertebral arteries
are usually caused by trauma (Kalimo et al., 2002).
Some 60% of saccular aneurysms are diagnosed
or found incidentally at post mortem in patients be-
tween 40 and 60 years of age. Less than 5% occur
under the age of 20 years (Kalimo et al., 2002). About
12–15% of saccular aneurysms are familial and occur
at a younger age than the sporadic aneurysms.
Saccular aneurysms show a restricted location with
85–90% arising on the terminal part of the internal
carotid artery or on the major branches of the anterior
portion of the circle of Willis (Fig. 2). The internal
carotid artery is the most frequent site (40%), followed
by the anterior communicating artery (30%) and the
proximal portion of the middle cerebral artery (20%).
In adults, 5–10% of aneurysms are associated with
the posterior cerebral or vertebral arteries whereas in
children 40–45% of aneurysms occur in the posterior
cerebral circulation (Kalimo et al., 2002). Multiple
saccular aneurysms occur in 10–31% of patients, most
frequently associated with the middle cerebral artery.
Saccular aneurysms occur on the feeding arteries of
arteriovenous malformations in 10% of cases (Batjer,
Suss, & Samson, 1986), adding support to the hypoth-
esis that haemodynamic stress plays a significant role
in the pathogenesis of aneurysms.
The majority of saccular aneurysms occur at or near
branches of major cerebral arteries around the cir-
cle of Willis. They vary in size from 3 to 4 mm to
the giant aneurysms ranging from 25 mm in diame-
ter to 4–5 cm in exceptional cases. Although many are
shaped like berries there is considerable variation in
shape and they are often multi-lobed. Some aneurysms
are almost completely destroyed during rupture and
subarachnoid haemorrhage, others have thin translu-
cent walls and a point of rupture at the apex. Many
aneurysms are thick walled and opaque. Variation in
shape not only applies to the fundus of the aneurysm
but also to the neck by which by which the aneurysm
arises from the parent artery. Some aneurysms have
very narrow necks whereas others have a broad neck
comparable with the diameter of the aneurysm itself.
Saccular aneurysms are distinguished from fusiform
aneurysms by their neck and not being in line with
the artery. Multiple aneurysms are notable in famil-
ial cases, with different epidemiological and clinical
characteristics (see Section 1.2); they can also oc-
cur in some sporadic cases. In the case of specific
gene defects, the aneurysmal characteristics may dif-
fer. In Ehlers-Danlos syndrome (EDS) type IV (see
Section 3.2), fusiform rather than saccular aneurysms
are observed (Mirza, Smith, & Lim, 1979).
Although saccular aneurysms may extend inferiorly
from the circle of Willis, some protrude superiorly into
the brain substance and may thus be associated more
with massive intracerebral or intraventricular haemor-
rhage rather than with subarachnoid haemorrhage.
All arteries and veins on the surface of the brain
are located in the subarachnoid space between the
thin outer layer of arachnoid mater and the delicate
 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
pia mater that coats the surface of the brain. When a
subarachnoid haemorrhage occurs due to rupture of
a saccular aneurysm, blood is extravasated at arterial
pressure into the subarachnoid space and spreads away
from the site of the ruptured aneurysm to cover large
areas of the surface of the brain. Filling the subarach-
noid space with arterial blood has a number of effects.
Delicate nerve fibres within the leptomeninges are
stimulated and may cause intense pain and headache.
Arteries within the subarachnoid space may undergo
spasm when surrounded by fresh blood and this re-
sults in poor cerebral perfusion and the ischaemia
may be so severe that infarction occurs in the areas
supplied by those arteries. Disruption of cerebrospinal
fluid flow within the subarachnoid space may oc-
cur both acutely and some years after subarachnoid
haemorrhage as the blood becomes organised into
scar tissue. As the scar tissue forms, it binds the outer
arachnoid mater to the pia mater and thus blocks the
subarachnoid space, impedes the drainage of cere-
brospinal fluid and thus hydrocephalus with dilatation
of the cerebral ventricles ensues (Heros, 1989). Other
structures within the subarachnoid space include the
cranial nerves that may be damaged by the subarach-
noid haemorrhage. Blood is confined to the subarach-
noid space and does not enter the perivascular spaces
around arteries in the brain as the subarachnoid space
is separated from perivascular spaces by the pia mater
(Hutchings & Weller, 1986; Weller, 1995).
Thus, the major complications of ruptured saccular
aneurysms are either due to intracerebral or intraven-
tricular haemorrhage or due to the effects of blood
within the subarachnoid space.
2.3. Pathogenesis and the stages involved in the
formation of saccular aneurysms
About 12–15% cases of saccular aneurysm are fa-
milial, often with an autosomal dominant pattern of
inheritance (Kalimo et al., 2002). There is an increased
incidence of saccular aneurysms in families with poly-
cystic kidney disease, fibromuscular dysplasia, and
moyamoya disease and other connective tissue disor-
ders including Marfan syndrome and EDS type IV.
Although originally hypertension was thought to be a
major factor in the formation of aneurysms, it appears
to play a minor role in the formation of aneurysms al-
though it may be associated with their rupture. Other
1347
factors such as cigarette smoking increase the risk of
subarachnoid haemorrhage although the mechanisms
are unclear (Fogelholm & Murros, 1987).
Some 85–90% of saccular aneurysms are sporadic
with no obvious predisposing systemic factors. In
these cases, it appears that haemodynamic stress at
branches of the circle of Willis is a major factor in
the pathogenesis of saccular aneurysms (Sheffield &
Weller, 1980; Weller, 1995). Branching patterns of
the circle of Willis show wide individual variation.
In many cases, the branches are asymmetrical. The
majority of aneurysms occur between the ages of 40
and 70 and this has highlighted the probability that
most sporadic aneurysms are due to ageing effects on
the arteries.
Cerebral arteries in infants are vessels with a rela-
tively thin smooth muscle coat (media) and a lumen
lined by endothelium abutting on to an internal elas-
tic lamina. There is little connective tissue separating
internal elastic lamina from the endothelium and the
internal elastic lamina itself is convoluted when ex-
amined at post mortem, as reflects the elasticity of the
artery wall. There is probably minimal turbulance at
the bifurcation (Fig. 3a).
With ageing, and probably due to haemodynamic
stress upon the endothelium, layers of fibrous tissue
develop between the endothelium and the internal
elastic lamina (Mordant & Weller 2002, unpublished
observation). This decreases the elasticity of the ves-
sel wall and furthermore raises mounds or pads of
fibrous tissue under the endothelium. Such pads are
particularly prominent at vessel bifurcations thus
gradually the sharp V-shaped bifurcation becomes
rounded (Fig. 3b). This change in shape of the bifur-
cation may alter the direction of haemodynamic stress
on the bifurcation resulting in pressure to form an out-
pouching at the bifurcation (Fig. 3c). It has long been
observed that the smooth muscle coat of arteries and
even the internal elastic lamina may be defective at the
bifurcations of cerebral arteries and this may be one
feature that predisposes these regions of an artery to
the formation of aneurysms (Fig. 4) (Fujimoto, 1996).
On microscopic examination of saccular aneurysms,
the smooth muscle wall (media) and the internal elastic
lamina of the artery end abruptly at the neck of the
aneurysm. The wall of the aneurysm itself is composed
of dense fibrous tissue of varying thickness. At the
apex of an aneurysm the fibrous tissue may be very
1348 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360

Fig. 3. Pathogenesis of saccular aneurysms: (a) in an infant artery there would be little turbulence of blood flow past the V-shaped carina
of the artery bifurcation; (b) haemodynamic stress damages the endothelium, pads of fibrous scar tissue form and change the conformation
of the vessel carina; (c) with the change in the direction of the haemodynamic stress, there is out-pouching of an aneurysm at the carina.

thin and this is the preferential site of rupture. In larger
aneurysms, layered thrombus may be formed on the
inner aspect of the aneurysm wall and there may be
little residual lumen.
Although much is known about the structure and
fate of saccular aneurysms and the evolution of their
structure can be presumed, the role of individual
components such as endothelium, fibrous tissue, in-
ternal elastic lamina and smooth muscle cells in the
formation of aneurysms is unclear. As pads of fibrous
tissue and aneurysms in cerebral vessels develop very
slowly it is rare to glimpse any of the intermediate

Fig. 4. Variations in the structure of cerebral artery bifurcations. In a proportion of arteries, there are gaps in the smooth muscle media
and internal elastic lamina.
 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360 1349

Fig. 5. Stages in the formation of subintimal fibrous pads at vessel bifurcations: (a) damage to the endothelium is followed by; (b) adhesion
of platelets and fibrin; (c) macrophages invade the thrombus on the surface of the vessel removing platelets and fibrin and this is followed
by; (d) Migration of fibroblasts into the area, the formation of a collagenous scar and re-growth of the endothelium. The protruding scar
tissue will increase turbulence of the blood flow and thus induce more endothelial damage. This repeated cycle of damage and scarring
increases the size of the pads around vessel bifurcations.

tissue changes that occur in the vessel wall during the
development of the aneurysm.
Fig. 5 sets out the predicted stages in the formation
of a saccular aneurysm drawing information from the
homologous tissue reactions that occur during the for-
mation of atherosclerotic plaques in arteries in many
regions of the body. Fig. 5a shows damage to the en-
dothelium from haemodynamic stress which results in
the deposition of platelets and fibrin on the surface
of the vessel wall (Fig. 5b). Monocyte–macrophages
from the blood or vessel wall remove the thrombus
on the wall of the artery (Fig. 5c) and as (Fig. 5d) fi-
broblasts form collagenous scar tissue of the small in-
timal pad, the endothelium grows over the surface to
reconstitute the vessel wall lining. Repeated episodes
of damage and scarring ensure the enlargement of in-
timal pads and the changes of shape of the vessel
wall leading to the haemodynamic forces extruding the
aneurysmal sac at the vessel bifurcation.Genes encod-
ing the proteins and enzymes involved in the cellular
pathogenesis of saccular aneurysms offer themselves
as possible candidates for the molecular pathogenesis
of subarachnoid haemorrhage. Although hypertension
appears to be a significant risk factor for the rupture
of saccular aneurysms, no clear association with struc-
tural components of the artery is apparent other than
the increased stress on the vessel wall itself (Coutard,
1999; Peters et al., 2001).
2.4. Cellular and molecular architecture of the
vessel wall
The cerebral arteries lack an external elastic lam-
ina but are otherwise similar to other arteries. Arteries
comprise sequentially from the lumen: a monolayer of
lining endothelial cells; an intima bounded by the in-
ternal elastic lamina; a medial layer; an external elas-
tic lamina; and adventitia. A cellular and molecular
overview is schematised in Fig. 6.
The elastic laminae mainly have a longitudinal ar-
rangement of fibres. These structures are responsible
for dilatation and recoil. The fibres contain an elastin
core formed from the polymerisation of tropoelastin
molecules which have a random coil structure but
which are cross-linked by lysyl oxidase. Lysyl ox-
idase, a Cu(II)-dependent enzyme, cross-links four
lysine sidechains to form the unique desmosine moi-
ety. These sidechains may belong to up to four dif-
ferent polypeptide chains, including those of other
arterial wall proteins such as fibrillins and colla-
gens. Within the elastic lamina, the elastin core is
modelled onto a microfibril lattice, itself made from
fibrillin-1 and -2. These are structurally very similar,
but display different temporal and spatial expression
patterns (Zhang et al., 1994; Zhang, Hu, & Ramirez,
1995). The main content of fibrillins is tandem re-
peats of calcium-binding epidermal growth factor-like
1350 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
domains interspersed by cysteine-rich sequences ho-
mologous to a TGF-␤ binding protein motif. Fib-
rillin monomers polymerise head-to-tail to form
microfibrils. Both linear, lateral and three-dimensional
interactions are stabilised by calcium (Kielty &
Shuttleworth, 1995). The magnitude and direction of
stresses on the vessel determines the shaping of the
elastic fibres. Microfibrils are distributed throughout
the arterial wall as an architecture of flexible links,
whereas the elastic fibres are longitudinal in the in-
ternal elastic lamina of the cerebral artery but radial
and concentric in the tunica media.
The other major extracellular constituent of the ar-
terial wall is the collagenous network, which contains
types I and III collagen. Type III collagen is specif-
ically associated with vascular diseases (see further)
whereas type I collagen is much more generally ex-
pressed. Type III fibres (reticular fibres) seem to form
particularly in tissues subjected to periodic stress. Type
III collagen is a homotrimer of ␣ subunits. The three
subunits nucleate self-assembly from their C-termini.
Critical elements to stable folding and assembly in-
clude the occurrence of glycine at every third amino
acid position and the hydroxylation of lysines and pro-
lines occurring in the positions immediately preceding
glycines in the sequence (van der & Garrone, 1991).
Fig. 6. Schematic of the layers, cellular and molecular constituents
of the cerebral arterial wall. EC, endothelial cell; IEL, internal
elastic lamina; SMC, smooth muscle cell.
A range of proteases (e.g. matrix metallopro-
teinases, elastase) and their inhibitors (TIMPs,
␣1-antitrypsin, ␣2-macroglobulin), growth factors
and cytokines determine the modelling and turnover
of the vascular wall, secreted by or regulating the
smooth muscle cells abundant in the tunica media,
or introduced by monocytes, macrophages and poly-
morphs arriving at the endothelial surface and fibrob-
lasts particularly in the adventitial layer responsible
for tensile strength.
Both structure of individual components, their inter-
actions and turnover are equally important to the over-
all integrity of the vessel wall and in this context both
environmental factors and genetic factors can exert di-
verse effects. It is easy to imagine, for example, how a
haploinsufficiency or defective protein (dominant neg-
ative effect) could lead to aneurysm, but less obvious
how that could lead to stenosis. However, considering
the interdependences of remodelling, a deficiency of
one component can lead to an overexpression of an-
other, as illustrated (see further) by elastin deficiency
in supravalvular aortic stenosis.
3. Molecular pathology
Several genes and their cognate proteins have been
specifically implicated in the molecular pathology of
aneurysms (Table 1) and these are considered in detail
in the following Sections 3.1–3.6.
3.1. PKD1 gene and polycystin
Adult polycystic kidney disease is an autosomal
dominant disorder in which renal cysts form, leading
to progressive loss of glomerular filtration and subse-
quently to renal failure and end-stage renal disease. As
early as 1960, Ditlefsen and Tonjum (1960) described
a family containing 15 cases with polycystic kidney
disease including 6 who had suffered from cere-
bral haemorrhage including one case who had been
proven to have a middle cerebral artery aneurysm.
Subsequent surveys (e.g. Chapman et al., 1992) have
contrasted the frequency of asymptomatic intracranial
aneurysms in polycystic kidney disease (upper 95%
confidence bound 9%), with that in the general popu-
lation (1%), although the lower 95% confidence bound
makes it possible that there is no difference from
 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360 1351

Table 1
Proteins of potential importance in intracranial aneurysm
Protein Evidence

Polycystin Mutated in polycystic kidney disease, in which IA seems to be increased

Collagen III Obvious candidate on account of the structural role of type III collagen in the arterial
wall; deficient in some cases of SAH; gene defects cause Ehlers-Danlos syndrome type
IV which predisposes arterial ruptures; similar outcomes for mouse gene knockout
Elastin Obvious candidate on account of role of elastic laminae in arterial integrity;
hypothesis-free linkage tests of genome in sib pairs affected by IA, show linkage to
genome region containing elastin gene—also strong association of IA with a specific
elastin haplotype in sib pairs
␣1-Antitrypsin (protease inhibitor) Major inhibitor of elastase, therefore obvious candidate; suggestion in case studies of
association of deficiency genotypes (of alleles Z and S) with IA/SAH
Collagen IV Intimal basement membrane component; mutated in Alport’s syndrome; one case
report of IA prior to traumatic SAH in 14-year-old boy in Alport’s syndrome
Fibrillin-1 Structural role in arterial wall microfibrils; mutated in Marfan syndrome of which a
major cause of mortality is abdominal aortic aneurysm and rupture; unproven role in IA
Collagen I Structural role in cerebral artery, but wide tissue expression profile
Fibrillin-2 Close homology and co-function with fibrillin-1
Matrix metalloproteinases, elastase, inhibitors Functional role in vessel wall turnover

population risk. Nevertheless, there can be no doubt
that in individual families with specific polycystic
kidney disease mutations, intracranial aneurysms and
haemorrhage can occur very frequently. The pleitropy
of the disease is further marked by other extrarenal
manifestations including cysts in the gastrointestinal
system and diverse cardiac vavular defects. Hossack,
Leddy, Johnson, Schrier, and Gabow (1988) con-
cluded that the latter defects marked a fundamental
biochemical defect in the extracellular matrix. The
gene causing adult polycystic kidney disease, named
PKD1, was mapped to an extremely CpG-rich re-
gion in human chromosome 16p13.3 (Germino et al.,
1992) and its complete structure was reported by the
International Polycystic Kidney Disease Consortium
in 1995. The 14.5 kb transcript comprises 46 exons
encoding a 4304-amino acid protein named poly-
cystin. The N-terminal half of the protein contains
leucine-rich repeats flanked by cysteine-rich struc-
tures, LDL-A and C-type lectin domains and fourteen
80-amino acid units of an unknown type of domain.
These domains suggest that polycystin participates in
protein–protein and protein–carbohydrate interactions
in the extracellular matrix. Immunostaining shows the
presence of polycystin in arterial smooth muscle cells
and myofibroblasts; and in intracranial aneurysms in
ADPKD there is disruption of elastic laminae (Griffin,
Torres, Grande, & Kumar, 1997). Polycystin 1, to-
gether with polycystin 2, is thought to function as a
part of a multiprotein membrane-spanning complex
involved in cell–cell and cell–matrix interactions.
Polycystin 1 can initiate signal transduction, leading
to the activation of various downstream effectors in-
cluding heterotrimeric G-proteins, protein kinase C,
mitogen-activated protein kinases, ␤-catenin and the
AP-1 transcription factor. At least in the renal con-
text, PKD cells respond to raised cyclic AMP by a
paradoxical increase in the rate of cell proliferation
(Calvet & Grantham, 2002). Reeders (1992) proposed
a two-hit mutational model for PKD1, a model well
recognised in the field of autosomal dominant can-
cers. Since most nephrons appear to be normal, it
could be that only a somatic mutation of the second
PKD1 allele in cells already affected by a germ line
defect of one PKD1 allele, leads to cyst formation.
Such a two-hit model should also be considered for
the extrarenal manifestations. Indeed, for autosomal
dominant forms of aneurysms in general (see fur-
ther), a two-hit hypothesis represents one possible
aetiological model. Mutations causing different de-
fects in the same protein are another aspect of the
genotype–phenotype relationship and of phenotypic
heterogeneity. However, for the PKD1 locus, there
does not seem to be any phenotypic difference of re-
nal features between patients with major PKD1 gene
deletions and those with single missense amino acid
1352 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
changes. This suggests that haploinsufficiency rather
than dominant negative effects occur. However, there
must be other modifiers since the same mutation in
different cases in the same family can manifest both
severe childhood and typical adult onset, as evidenced
in a family with a tyr3818stop mutation (Peral et al.,
1996). Mutations specifically predisposing intracra-
nial aneurysms have not been reported.
3.2. COL3A1 and type III collagen
Foetal and blood vessel collagen is also called colla-
gen III and it has long been known that its synthesis is
defective in EDS type IV. The COL3A1 gene on chro-
mosome 2 (2q31) encodes the polypeptide known to be
defective in this syndrome. The features of EDS type
IV (arterial rupture risk, thin transparent skin, easy
bruising, joint laxity, ligament weakness, bowel rup-
ture and other soft tissue risks) and vascular expression
of COL3A1 render this gene an obvious candidate in
all types of arterial aneurysms. Inactivation of the gene
in the mouse results in much shorter life span, with the
main cause of death being rupture of major blood ves-
sels (Liu, Wu, Byrne, Krane, & Jaenisch, 1997). Sys-
tematic study of many different mutations identified
in EDS type IV has not revealed genotype–phenotype
correlation. Nevertheless, vascular events without
other features have been observed in conjunction with
COL3A1 mutation, for example Gly619Arg heterozy-
gosity was identified in a family prone to fatal rupture
of aortic aneurysm. This particular mutation causes
type III collagen to unfold at a decreased temperature.
As early as 1981, study of type III collagen in 12 pa-
tients with cerebral aneurysms revealed deficiency in 7
(Pope et al., 1981). Similar studies by others have sub-
sequently confirmed these observations (Ostergaard
& Oxlund, 1987). Subsequent studies by direct muta-
tion scanning in sporadic cases and cases with at least
one affected relative, have not identified COL3A1
mutations (van den Berg et al., 1999). However, both
the nature of the gene (collagen glycine repeating se-
quences may perform atypically in mutation scanning
techniques); the case ascertainment (even the presence
of one affected relative may only be weak evidence
of strong familiality); the scanning techniques so far
used; and the focus of attention solely on coding
regions, do not exclude the possibility that the gene
could be a significant cause of familial occurrences
of SAH in which there is a single major gene effect.
Less attention has been paid to COL3A1 as a minor
risk gene, for example examining common haplotypes
by association in sporadic case population or trios de-
signs, although one small study has claimed a signif-
icant association between an Ava II site and sporadic
cerebral aneurysms (Brega, Seltzer, Munro, & Breeze,
1996). More recently, 2 of 16 patients with cervical
arterial dissections were shown to have a low ratio of
type III to type I collagen, although mutations could
not be detected in COL3A1 (van den Berg et al., 1998).
3.3. Fibrillin-1 and other linkages in aortic and
mixed vessel aneurysmal disease
The Marfan syndrome is an autosomal dominant
heritable disorder of connective tissue with prominent
manifestations affecting the skeletal, ocular and car-
diovascular systems. The incidence is estimated to be
up to 1/5000 (Pyeritz, 2000) with perhaps 25% of cases
representing new mutations. Essentially all classical
Marfan families and cases have turned out to display
linkage to human chromosome 15q21.1 and/or muta-
tions in the FBN1 gene encoding fibrillin-1. Mutations
in this same gene also cause a series of other related
disorders of connective tissue collectively known as
type I fibrillinopathies (Robinson et al., 2002). Whilst
the skeletal abnormalities of Marfan syndrome may
be the most obvious, including long bone overgrowth,
joint laxity, scoliosis and pectus abnormalities, it is
the cardiovascular features which are often fatal. In
particular, progressive dilatation of the aortic root and
aortic dissection and rupture are frequent, and mitral
and aotic valve insufficiency may also occur early.
These features are the predominant causes of death
in over 90% of patients (Nienaber & Von Kodolitsch,
1999). Genotype–phenotype correlations are recog-
nised and neonatal onset Marfan syndrome associates
particularly with mutations in exons 24–32. Such sub-
jects show unusual features including crumpled ears,
flexion contractures, pulmonary emphysema and loose
skin, and often die before age 2 years from cardiac
failure caused by valve defects (Booms et al., 1999).
Whereas one-third of Marfan mutations are premature
stop codons, splice and exon deletions, no nonsense
mutation has been described for the neonatal disease,
in which mutations identified are either amino acid
changes or exon skipping leading to shortened fibillin
 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
monomers taking part in fibril formation. This sug-
gests that there might be dominant negative effect in
the neonatal subclass of Marfan syndrome, whereas
haploinsufficiency is the more likely basis of effect
where one allele is not expressed at all. Mutation
R1170H, on the other hand, has been observed in two
families with mild features and lacking aortic compli-
cations. A distinct set of missense mutations in exons
24–32 seem to predispose to early onset of cardio-
vascular complications (Putnam et al., 1996). Milder
fibrillinopathies associated with isolated ectopia lentis
(e.g. Lonnqvist et al., 1994), isolated skeletal defects
and other combinations, show mutations clustering in
exons 59–65. Missense mutations in the proline-rich
exons 1–10 seem to lead to late-onset, mild cardio-
vascular complications. No study has been reported
of the fibrillin-1 gene in other categories of vascular
dilatation and aneurysm, but it is clear that such stud-
ies will be well worthwhile, although laborious, given
the pleitropy and phenotypic heterogeneity already ob-
served for fibrillin-1 gene variations.
While there is little family-based linkage data for
SAH yet, more linkage information has begun to
emerge for aortic aneurysms. In addition to recog-
nition of the role of fibrillin-1 in aortic aneurysm,
studies during 2001 have identified linkage of familial
aortic aneurysm (FAA) to chromosome 11q23.2–q24
in one family and excluded linkage to fibrillin-1, to
COL3A1, to 11q, to 3p24.2–p25, or fibrillin-2 in an-
other family (Vaughan et al., 2001). A study of 15
families with dominant inheritance of thoracic aortic
aneurysms (TAA) and dissections revealed a linkage
in 9 families to chromosome 5q13–q14 (Guo et al.,
2001). The loci on chromosomes 5 and 11 have not
yet been identified but clearly represent protein prod-
ucts and pathways which may also be relevant to
intracranial aneurysms. Within the Wessex region of
the UK, we are undertaking studies of a family with
some Marfan features including aortic pathology in
several members, and several occurrences of intracra-
nial aneurysm or bleed have occurred in the same kin-
dred (Dennis, Simpson, Day, & Zhang, unpublished
observation). The range of locus heterogeneity, gene
pleitropy and phenotypic heterogeneity of specific
mutations first requires the identification of each of
the specific biochemical lesions, but it is evident that
there will be some overlap of pathogenesis between
some categories of arterial aneurysm.
1353
3.4. Genomewide-linkage and haplotype association
at chromosome 7q11 in the elastin gene
Traditional genetic linkage studies rely on the iden-
tification and sampling of a large family in which a
disease is segregating in a clearly bimodal fashion.
Characterisation of several hundred highly polymor-
phic microsatellite loci distributed evenly throughout
the genome, in all family members, enables the iden-
tification of which region of which chromosome is
segregating with the disease. This usually narrows
down the search for the relevant gene and its protein
product, to a choice between tens or hundreds of
‘positional candidates’ within the linked region. The
recent availability of a reasonably complete human
genome sequence (Lander et al., 2001) has facilitated
the identification of positional candidates, although
frequently the investigator must still first improve the
map or sequence quality in the region. Unfortunately,
although kindreds with many occurrences of SAH are
known, the frequent fatality of the condition at first
presentation has hindered DNA sampling of sufficient
family members (e.g. 10 affected members) to obtain
statistical power in any single family. Furthermore,
although more recently imaging techniques have en-
abled identification of a pre-symptomatic trait, routine
screening or research screening are ethically complex
given the uncertain benefit relative to the risks, of
preventative interventions. An alternative approach to
single large kindred studies, is to pool resources from
kindreds containing at least two affected members. In
this non-parametric design (i.e. no assumptions about
dominance or recessivity of inheritance), chromosome
regions will, according to Mendelian segregation, dis-
play sharing of the same pair of parental alleles for
25% of first degree relatives; sharing of one parental
allele for 50% of first degree relatives; and sharing
of no parental allele for 25% of first degree rela-
tives. However, if a genomic region is important to
a disease, and relative pairs were ascertained on the
basis of their dual affectation status, then observed
allele sharing for that region will be greater than that
expected. This approach to analysis does not require
that a region be causative, it could just be predispos-
ing: neither does the approach require that the region
contains a gene acting in all families. However, the
combination of extent of effect and number of fam-
ilies in which the gene is acting, must be sufficient
1354 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
to be statistically evident in the eventual analysis.
The approach should therefore be successful in any
disease for which there is a substantial genetic com-
ponent caused by a relatively small number of loci
(genes) overall in the population. This approach has
recently been applied in Japan, in a study comprising
104 sib pairs affected by intracranial aneurysm, col-
lected from systematic enquiry in 1100 neurosurgical
centres (Onda et al., 2001). Genomewide, the three
most prominent linkage regions identified were 7q11,
14q22 and 5q22–q31, with maximum LOD scores
of 3.22, 2.31 and 2.24, respectively. Notwithstanding
issues of multiple statistical testing, expected number
of positive regions and multipoint versus single point
linkage, these values correspond with P-values less
than 0.001, 0.01 and 0.01, respectively. Linkage on
7q11 was strongest at microsatellite locus D7S2472.
Proximity to the elastin gene (ELN) was notable, and
so further analysis of this gene was undertaken using
a set of common single nucleotide polymorphisms to
define common haplotypes. Association analysis iden-
tified a haplotype comprising markers in introns 20
and 23 which was more prevalent in aneurysm patients
than in controls, at an odds ratio of 1.85. In a global
test of all pairwise combinations from nine polymor-
phisms, this haplotype emerged with P = 3.81×10−6
(Onda et al., 2001). Candidate genes in the chromo-
some 5 region include lysyl oxidase (LOX), fibrillin-2
(FBN2) and fibroblast growth factor 1 (FGF1); and
latent transforming ␤-binding protein 2 (LTBP2) on
chromosome 14. While external elastic lamina is
absent in cerebral arteries, internal elastic lamina
(Fig. 6) is the major structural support (Glynn, 1940).
Pathological examinations of intracranial aneurysms
have pointed to defects of the internal elastic lam-
ina (Carmichael, 1945). Elastin fibril destruction by
elastase (Miskolczi, Guterman, Flaherty, & Hopkins,
1998) or by ␤-aminopropionitrile which inhibits
elastin cross-linking (Hashimoto, Handa, & Hazama,
1978) both cause intracranial aneurysm formation in
animal models. The statistical approach used to impli-
cate the elastin locus (or plausibly not the elastin locus
but a nearby locus in strong linkage disequilibrium)
seems have yielded significant evidence that variation
in the elastin gene affecting either its molecular struc-
ture or expression level, may be important in intracra-
nial aneurysm. However, a combination of replication
and functional follow studies will now be needed to
confirm and further clarify the possible aetiological
chain.
Genotype–phenotype considerations may be im-
portant for the elastin gene. It was shown by Curran
et al. (1993) that a breakpoint in the elastin gene in
7q11.23 causes supravalvular aortic stenosis. Other
studies Ewart et al. (1993) showed that this same
region is deleted in Williams syndrome in which
supravalvular aortic stenosis is also a major feature.
All mutations identified (deletions, splice and stop
codons) are consistent with a haploinsufficiency ef-
fect, although the penetrance of the effect can vary
between members of the same family. The basis of
arterial thickening occurring as a result of elastin de-
ficiency has been elucidated by mouse gene knockout
experiments by Li et al. (1998). Deficiency of elastin
led to subendothelial cell proliferation and reorgani-
sation of smooth muscle. Some frameshift mutations
of elastin have been observed as the basis of cutis
laxa, a condition of loose skin, also with tendency to
hernias, valvular and cardiac abnormalities (Zhang
et al., 1999). These observations would suggest that
if elastin is an aetiological gene in aneurysm forma-
tion, then either dominant negative (e.g. by a changed
amino acid) rather than haploinsufficiency effect must
be the cause, or alternately that tissue-specific expres-
sion, splicing, matrix structure or other feature enable
the elastin gene to lead to vessel thickening in one
context but aneurysm formation in another.
3.5. Collagen IV and other basement membrane
constituents
While there have been few specific studies of other
proteins in the wall of the cerebral artery, several pro-
tein families recognised in other arteries merit further
investigation. The intimal layer (Fig. 6) contains a
basement membrane which supports the endothelial
cells. Other basement membranes contain other spe-
cialised collagens such as collagen IV and collagen
XVIII. Collagen XVIII trimerises via its carboxyl-
terminal domain which as a cleavage product be-
comes endostatin. Endostatin domain oligomerisation
is essential for inhibition of endothelial capillary tube
formation in vitro and the more diffusible endostatin
localises to the elastic lamina. Two separate sites
on endostatin respectively bind cell surface heparin
sulphate glycosaminoglycan and laminin (Javaherian
 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
et al., 2002). Laminins, of which there are several
distinct genes and polypeptide chains, are found in
both the intima and medial layers of the artery. An-
other basement membrane constituent interacting
with laminin, is nidogen. Both these proteins and
fibronectin and collagen IV display remodelling in ex-
perimentally formed aneurysms in response to haemo-
dynamic stresses (Kittelberger, Davis, & Stehbens,
1990). Collagen IV has been shown to be present in
cerebral artery basement membranes. Molecular de-
fects in collagen IV cause Alport’s syndrome, which
is characterised by sensorineural hearing loss and
haematuria. However, a recent case report of a trau-
matically induced subarachnoid haemorrhage in a
14-year-old male identified a pre-existing left carotid
artery bifurcation aneurysm and a pre-existing his-
tory, proven by renal biopsy, of Alport’s syndrome
(Vaicys, Hunt, & Heary, 2000). It is therefore a plau-
sible new hypothesis that defects in the basement
membrane as well as defects in the elastic and medial
layers may predispose intracranial aneurysms. It is
also an example which invites further investigation of
the possibility that traumatic subarachnoid haemor-
rhage could be more frequent in individuals rendered
susceptible by molecular structural variations leading
to a more fragile arterial wall.
3.6. α1-Antitrypsin (protease inhibitor)
Despite its name, ␣1-antitrypsin is the most
abundant and potent natural inhibitor of elastase.
Deficiency is a well known cause of pulmonary em-
physema and liver disease, particularly associated with
homozygosity for the ‘Z’ allele which leads to reduced
serum levels and enzyme function and particularly ex-
acerbated in smokers. ␣1-Antitrypsin activity has been
claimed to be reduced in both intracranial (Tartara
et al., 1996) and abdominal aortic (Cohen, Mandell,
Margolis, Chang, & Wise, 1987) aneurysm. The epi-
demiological risk of SAH in smokers (see Section
3.7) is also of note. Some relatively small studies of
␣1-antitrypsin genotype in sporadic SAH case series,
classifying wild type allele (M), deficient allele (Z) and
slightly deficient allele (S), have claimed positive asso-
ciation of S and Z heterozygosity with SAH, for exam-
ple Schievink, Katzmann, Piepgras, and Schaid (1996)
finding 16% prevalence for 100 consecutive cases
compared with 8% population prevalence. In another
1355
study, St. Jean et al. (1996) found an 8-fold overrepre-
sentation of the Z allele in a subsample of 46 intracra-
nial aneurysm patients, but in the overall study and
after correction for multiple testing, this was not sig-
nificant. Larger study sizes or eventual meta-analysis
of literature will be necessary to determine whether the
existent literature represents a positive reporting bias.
3.7. Association studies in sporadic intracranial
aneurysm and subarachnoid haemorrhage
First degree relatives of patients with subarachnoid
haemorrhage recruited prospectively on presentation,
show an estimated 2–5% lifetime risk of also develop-
ing a subarachnoid haemorrhage (Bromberg, Rinkel,
Algra, Greebe, et al., 1995). This suggests that there
may be genetic modifiers in sporadic intracranial
aneurysm and haemorrhage in addition to the clear-cut
major gene effects in a small subset of families. Stud-
ies of common genetic diversity which might exert
such effects, have followed similar hypotheses to
those applied to abdominal aortic aneurysm. Studies
are focussing on several categories of gene, mainly
in the context of causation but occasionally in the
context of outcome:
1. Genes encoding matrix proteins of the vessel wall,
for example common variation of the COL3A1 gene
(Brega et al., 1996).
2. Genes encoding matrix modifiers (e.g. enzymes) of
the vessel wall (Peters, Kassam, St. Jean, Yonas,
& Ferrell, 1999; Zhang et al., 2001), genes encod-
ing cytokines and other factors which may modify
cellular function.
3. Classic genetic epidemiological candidate genes,
HLA, apolipoprotein E gene (APOE); and an-
giotensin converting enzyme gene (ACE) (Dunn,
Stewart, Murray, Nicoll, & Teasdale, 2001; Hirose
et al., 1998; Keramatipour et al., 2000; Leung,
Poon, Yu, Wong, & Ng, 2002; Niskakangas et al.,
2001; Ryba et al., 1992; Takenaka et al., 1998).
4. Genes influencing atherothrombotic and oxidation
pathways such as APOE, lipoprotein(a) (Roberts
et al., 2001), and methylene tetrahydrofolate reduc-
tase (MTHFR).
Broadly, the same candidates have been consid-
ered in intracranial aneurysm events as in abdomi-
nal aortic aneurysm formation and rupture. However,
1356 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
no claim has been consistently replicated with strong
statistical significance. A specific difficulty facing re-
search groups is that sporadic subarachnoid haemor-
rhage is rare and pre-symtomatic aneurysms cannot
readily be ascertained. Thus most studies in regional
centres based on sequential case admissions (and suit-
ably selected controls) are based on study sizes of
50–200 subjects, which limits power to detect any
genotypic effect of less than a two–three-fold relative
risk (Risch & Merikangas, 1996). However, such mag-
nitude of effect would represent a comparatively major
modifier gene, not dissimilar from the effects of HLA
genotype in type I diabetes (Undlien, Lie, & Thorsby,
2001), of APOE alleles in late onset Alzheimer’s dis-
ease occurrence (Saunders, 2000), and greater than the
odds ratio of the specific haplotype identified in the
elastin gene apparently conferring an odds ratio of 1.85
for intracranial aneurysm in patients selected on the
basis of being a member of a family containing two af-
fected sibs (Onda et al., 2001). Ultimately therefore, as
studies accrue, literature meta-analysis or prior pool-
ing of clinical collections will be necessary to fully
explore gene effects in sporadic intracranial aneurysm.
4. Future lines of investigation
The target of molecular investigations in sub-
arachnoid haemorrhage is to understand the common
sporadic category, but there is some promise that the
smaller subset of truly familial occurrence will pro-
vide the tools to gain a greater insight. Large affected
relative pair studies and long term prospective sam-
pling and follow up to achieve single kindred linkage
studies will be essential. Animal models using gene
knockouts, conditional knockouts or specific intro-
duced mutations will help to determine sequences of
pathogenesis. Comprehensive expression studies us-
ing microarray technology may help to define more
fully, the expression changes taking place in cells
of the vessel wall during cerebral artery aneurysm
development. Genetic epidemiological studies may
identify gene–environment interactions of signifi-
cance. For example, smoking has an important role,
and the elastin genomic region has been implicated
by linkage, independent of any functional hypoth-
esis in aneurysm formation. Hypothesis testing has
also claimed association of ␣1-antitrypsin (elastase
inhibitor) genotype with subarachnoid aneurysm.
However, very large well designed epidemiological
analyses will be needed to prove the latter and test in-
teraction of the three factors. Parallel functional stud-
ies may be equally informative. Further systematic
association analysis in sporadic SAH of haplotypes
of genes encoding structural proteins of the arterial
wall and regulatory factors and enzymes, and high
throughput mutation scanning of these same genes in
families prone to SAH, may be informative.
Gene tests to reassure at risk family members
that they have not inherited a defective gene, and to
identify others for frequent non-invasive imaging and
safer minimally invasive management of ‘critical’
aneurysms, would be of clinical value. It is also pos-
sible that other avoidable environmental factors may
be identified during cellular and molecular studies
of pathogenesis. Drugs capable of modifying matrix
turnover are an active area of pharmaceutical research
and given identifiable, predictable subphenotypes of
SAH, could also plausibly become part of a future
arsenal of medical options.
Note added in proof
A second genomewide linkage scan, comparable
with the Japanese study described in Section 3.4,
has recently been published (Olson et al., BMC Med
Genet, 2002, 3(1) 7). This study of 85 Finnish fam-
ilies identified several possible linkages, the most
significant (maximal multipoint LOD score 2.6) be-
ing on chromosome 19q. The possible casual gene in
this interval is not obvious although there are many
plausible candidates to be investigated. The general
lack of correlation of linkages with the Japanese study
may reflect occurrence of different risk genotypes and
different molecular causality of familial SAH in these
very different populations.
Acknowledgements
BZ is a Hope (Wessex Medical Trust) Senior
Research Fellow and thanks the British Heart Foun-
dation for project grant support. INMD thanks Hope
for their pilot support to develop the Southampton
Familial Subarachnoid Haemorrhage project, research
sister Lesley Foulkes for family tracing and the late
 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
Professor Fausto Iannotti for collaboration in the early
stages of the programme.
References
Batjer, H., Suss, R. A., & Samson, D. (1986). Intracranial arterio-
venous malformations associated with aneurysms. Neuro-
surgery, 18, 29–35.
Bhattacharya, J. J., Luo, C. B., Suh, D. C., Alvarez, H., Rodesch,
G., & Lasjaunias, P. (2001). Wyburn-Mason or Bonnet-
Dechaume-Blanc as cerebrofacial arteriovenous metameric
syndromes (CAMS). Interventional Neuroradiology, 7, 5–17.
Booms, P., Cisler, J., Mathews, K. R., Godfrey, M., Tiecke, F.,
Kaufmann, U. C., Vetter, U., Hagemeier, C., & Robinson, P. N.
(1999). Novel exon skipping mutation in the fibrillin-1 gene:
Two ‘hot spots’ for the neonatal Marfan syndrome. Clinical
Genetics, 55, 110–117.
Brega, K. E., Seltzer, W. K., Munro, L. G., & Breeze, R. E.
(1996). Genotypic variations of type III collagen in patients
with cerebral aneurysms. Surgical Neurology, 46, 253–256.
Bromberg, J. E., Rinkel, G. J., Algra, A., Greebe, P., van Duyn,
C. M., Hasan, D., Limburg, M., ter Berg, H. W., Wijdicks,
E. F., & van Gijn, J. (1995). Subarachnoid haemorrhage in
first and second degree relatives of patients with subarachnoid
haemorrhage. British Medical Journal, 311, 288–289.
Bromberg, J. E., Rinkel, G. J., Algra, A., Limburg, M., & van
Gijn, J. (1995). Outcome in familial subarachnoid hemorrhage.
Stroke, 26, 961–963.
Bromberg, J. E., Rinkel, G. J., Algra, A., van Duyn, C. M., Greebe,
P., Ramos, L. M., & van Gijn, J. (1995). Familial subarachnoid
hemorrhage: Distinctive features and patterns of inheritance.
Annals of Neurology, 38, 929–934.
Calvet, J. P., & Grantham, J. J. (2002). The genetics and physiology
of polycystic kidney disease. Seminars in Nephrology, 21(2),
107–123.
Carmichael, R. (1945). Gross defects in the muscular and elastic
coats of the larger cerebral arteries. Journal of Pathology and
Bacteriology, 57, 345–351.
Chapman, A. B., Rubinstein, D., Hughes, R., Stears, J. C., Earnest,
M. P., Johnson, A. M., Gabow, P. A., & Kaehny, W. D. (1992).
Intracranial aneurysms in autosomal dominant polycystic kidney
disease. New England Journal of Medicine, 327, 916–920.
Cohen, J. R., Mandell, C., Margolis, I., Chang, J., & Wise, L.
(1987). Altered aortic protease and antiprotease activity in
patients with ruptured abdominal aortic aneurysms. Surgical
Gynecology and Obstetrics, 164, 355–358.
Coutard, M. (1999). Experimental cerebral aneurysms in the
female heterozygous Blotchy mouse. International Journal of
Experimental Pathology, 80, 357–367.
Curran, M. E., Atkinson, D. L., Ewart, A. K., Morris, C. A.,
Leppert, M. F., & Keating, M. T. (1993). The elastin gene is
disrupted by a translocation associated with supravalvular aortic
stenosis. Cell, 73, 159–168.
De Braekeleer, M., Perusse, L., Cantin, L., Bouchard, J. M.,
& Mathieu, J. (1996). A study of inbreeding and kinship in
intracranial aneurysms in the Saguenay Lac-Saint-Jean region
1357
(Quebec, Canada). Annals of Human Genetics, 60(Pt 2), 99–
104.
Ditlefsen, E. M. L., & Tonjum, A. M. (1960). Intracranial
aneurysms and polycystic kidneys. Acta Medica Scandinavica,
168, 51–54.
Dunn, L. T., Stewart, E., Murray, G. D., Nicoll, J. A., & Teasdale,
G. M. (2001). The influence of apolipoprotein E genotype
on outcome after spontaneous subarachnoid hemorrhage: A
preliminary study. Neurosurgery, 48, 1006–1010.
Etchevers, H. C., Vincent, C., Le Douarin, N. M., & Couly, G. F.
(2001). The cephalic neural crest provides pericytes and smooth
muscle cells to all blood vessels of the face and forebrain.
Development, 128, 1059–1068.
Ewart, A. K., Morris, C. A., Atkinson, D., Jin, W., Sternes, K.,
Spallone, P., Stock, A. D., Leppert, M., & Keating, M. T.
(1993). Hemizygosity at the elastin locus in a developmental
disorder, Williams syndrome. Nature Genetics, 5, 11–16.
Fogelholm, R., Hernesniemi, J., & Vapalahti, M. (1993). Impact
of early surgery on outcome after aneurysmal subarachnoid
hemorrhage. A population-based study. Stroke, 24, 1649–1654.
Fogelholm, R., & Murros, K. (1987). Cigarette smoking and
subarachnoid haemorrhage: A population-based case–control
study. Journal of Neurology, Neurosurgery and Psychiatry, 50,
78–80.
Frazee, J. G., Cahan, L. D., & Winter, J. (1980). Bacterial intra-
cranial aneurysms. Journal of Neurosurgery, 53(5), 633–641.
Fujimoto, K. (1996). ‘Medial defects’ in the prenatal human
cerebral arteries: An electron microscopic study. Stroke, 27,
706–708.
Gaist, D., Vaeth, M., Tsiropoulos, I., Christensen, K., Corder, E.,
Olsen, J., & Sorensen, H. T. (2000). Risk of subarachnoid
haemorrhage in first degree relatives of patients with
subarachnoid haemorrhage: Follow up study based on national
registries in Denmark. British Medical Journal, 320, 141–145.
Germino, G. G., Weinstat-Saslow, D., Himmelbauer, H., Gillespie,
G. A., Somlo, S., Wirth, B., Barton, N., Harris, K. L., Frischauf,
A. M., & Reeders, S. T. (1992). The gene for autosomal
dominant polycystic kidney disease lies in a 750-kb CpG-rich
region. Genomics, 13, 144–151.
Glynn, L. (1940). Medial defects in the circle of Willis and
their relation to aneurysm formation. Journal of Pathology and
Bacteriology, 51, 213–222.
Griffin, M. D., Torres, V. E., Grande, J. P., & Kumar, R. (1997).
Vascular expression of polycystin. Journal of the American
Society of Nephrology, 8, 616–626.
Guo, D., Hasham, S., Kuang, S. Q., Vaughan, C. J., Boerwinkle,
E., Chen, H., Abuelo, D., Dietz, H. C., Basson, C. T., Shete,
S. S., & Milewicz, D. M. (2001). Familial thoracic aortic
aneurysms and dissections: Genetic heterogeneity with a major
locus mapping to 5q13–q14. Circulation, 103, 2461–2468.
Hashimoto, N., Handa, H., & Hazama, F. (1978). Experimentally
induced cerebral aneurysms in rats. Surgical Neurology, 10,
3–8.
Heros, R. C. (1989). Acute hydrocephalus after subarachnoid
hemorrhage. Stroke, 20, 715–717.
Hirose, H., Takagi, M., Miyagawa, N., Hashiyada, H., Noguchi,
M., Tada, S., Kugimiya, T., & Tilson, M. D. (1998). Genetic
1358 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
risk factor for abdominal aortic aneurysm: HLA-DR2(15),
a Japanese study. Journal of Vascular Surgery, 27, 500–
503.
Horten, B. C., Abbott, G. F., & Porro, R. S. (1976). Fungal
aneurysms of intracranial vessels. Archives of Neurology, 33,
577–579.
Hossack, K. F., Leddy, C. L., Johnson, A. M., Schrier, R. W., &
Gabow, P. A. (1988). Echocardiographic findings in autosomal
dominant polycystic kidney disease. New England Journal of
Medicine, 319, 907–912.
Hutchings, M., & Weller, R. O. (1986). Anatomical relationships
of the pia mater to cerebral blood vessels in man. Journal of
Neurosurgery, 65, 316–325.
International Subarachnoid Aneurysm Trial (ISAT) Collaborative
Group (2002). International Subarachnoid Aneurysm Trial
(ISAT) of neurosurgical clipping versus endovascular coiling
in 2143 patients with ruptured intracranial aneurysms: A
randomised trial. Lancet, 360, 1267–1274.
Isaksen, J., Egge, A., Waterloo, K., Romner, B., & Ingebrigtsen, T.
(2002). Risk factors for aneurysmal subarachnoid haemorrhage:
The Tromso study. Journal of Neurology, Neurosurgery and
Psychiatry, 73, 185–187.
Javaherian, K., Park, S. Y., Pickl, W. F., LaMontagne, K. R., Tjin
Tham, S. R., Gillies, S., & Lo, K. M. (2002). Laminin modulates
morphogenic properties of the collagen XVIII endostatin
domain. The Journal of Biological Chemistry, 227(47), 45211–
45218.
Kalimo, H., Kaste, M., & Haltia, M. (2002). Vascular diseases. In
D. I. Graham, P. L. Lantos (Eds.), Greenfield’s neuropathology
(pp. 281–356) (7th ed.). London: Arnold.
Kasuya, H., Onda, H., Takeshita, M., Hori, T., & Takakura, K.
(2000). Clinical features of intracranial aneurysms in siblings.
Neurosurgery, 46, 1301–1305.
Keramatipour, M., McConnell, R. S., Kirkpatrick, P., Tebbs,
S., Furlong, R. A., & Rubinsztein, D. C. (2000). The
ACE I allele is associated with increased risk for ruptured
intracranial aneurysms. Journal of Medical Genetics, 37, 498–
500.
Kielty, C. M., & Shuttleworth, C. A. (1995). Fibrillin-containing
microfibrils: Structure and function in health and disease. The
International Journal of Biochemistry & Cell Biology, 27, 747–
760.
Kissela, B. M., Sauerbeck, L., Woo, D., Khoury, J., Carrozzella, J.,
Pancioli, A., Jauch, E., Moomaw, C. J., Shukla, R., Gebel, J.,
Fontaine, R., & Broderick, J. (2002). Subarachnoid hemorrhage:
A preventable disease with a heritable component. Stroke, 33,
1321–1326.
Kittelberger, R., Davis, P. F., & Stehbens, W. E. (1990).
Distribution of type IV collagen, laminin, nidogen and
fibronectin in the haemodynamically stressed vascular wall.
Histology and Histopathology, 5, 161–167.
Kojima, M., Nagasawa, S., Lee, Y. E., Takeichi, Y., Tsuda,
E., & Mabuchi, N. (1998). Asymptomatic familial cerebral
aneurysms. Neurosurgery, 43, 776–781.
Lander, E. S., Linton, L. M., Birren, B., Nusbaum, C., Zody,
M. C., Baldwin, J., Devon, K., Dewar, K., Doyle, M.,
FitzHugh, W., Funke, R., Gage, D., Harris, K., Heaford, A.,
Howland, J., Kann, L., Lehoczky, J., LeVine, R., McEwan,
P., McKernan, K., Meldrim, J., Mesirov, J. P., Miranda, C.,
Morris, W., Naylor, J., Raymond, C., Rosetti, M., Santos, R.,
Sheridan, A., Sougnez, C., Stange-Thomann, N., Stojanovic,
N., Subramanian, A., Wyman, D., Rogers, J., Sulston, J.,
Ainscough, R., Beck, S., Bentley, D., Burton, J., Clee, C.,
Carter, N., Coulson, A., Deadman, R., Deloukas, P., Dunham,
A., Dunham, I., Durbin, R., French, L., Grafham, D., Gregory,
S., Hubbard, T., Humphray, S., Hunt, A., Jones, M., Lloyd, C.,
McMurray, A., Matthews, L., Mercer, S., Milne, S., Mullikin,
J. C., Mungall, A., Plumb, R., Ross, M., Shownkeen, R., Sims,
S., Waterston, R. H., Wilson, R. K., Hillier, L. W., McPherson,
J. D., Marra, M. A., Mardis, E. R., Fulton, L. A., Chinwalla,
A. T., Pepin, K. H., Gish, W. R., Chissoe, S. L., Wendl, M.
C., Delehaunty, K. D., Miner, T. L., Delehaunty, A., Kramer,
J. B., Cook, L. L., Fulton, R. S., Johnson, D. L., Minx, P.
J., Clifton, S. W., Hawkins, T., Branscomb, E., Predki, P.,
Richardson, P., Wenning, S., Slezak, T., Doggett, N., Cheng,
J. F., Olsen, A., Lucas, S., Elkin, C., Uberbacher, E., Frazier,
M., Gibbs, R. A., Muzny, D. M., Scherer, S. E., Bouck, J. B.,
Sodergren, E. J., Worley, K. C., Rives, C. M., Gorrell, J. H.,
Metzker, M. L., Naylor, S. L., Kucherlapati, R. S., Nelson, D.
L., Weinstock, G. M., Sakaki, Y., Fujiyama, A., Hattori, M.,
Yada, T., Toyoda, A., Itoh, T., Kawagoe, C., Watanabe, H.,
Totoki, Y., Taylor, T., Weissenbach, J., Heilig, R., Saurin, W.,
Artiguenave, F., Brottier, P., Bruls, T., Pelletier, E., Robert, C.,
Wincker, P., Smith, D. R., Doucette-Stamm, L., Rubenfield, M.,
Weinstock, K., Lee, H. M., Dubois, J., Rosenthal, A., Platzer,
M., Nyakatura, G., Taudien, S., Rump, A., Yang, H., Yu, J.,
Wang, J., Huang, G., Gu, J., Hood, L., Rowen, L., Madan, A.,
Qin, S., Davis, R. W., Federspiel, N. A., Abola, A. P., Proctor,
M. J., Myers, R. M., Schmutz, J., Dickson, M., Grimwood, J.,
Cox, D. R., Olson, M. V., Kaul, R., Raymond, C., Shimizu, N.,
Kawasaki, K., Minoshima, S., Evans, G. A., Athanasiou, M.,
Schultz, R., Roe, B. A., Chen, F., Pan, H., Ramser, J., Lehrach,
H., Reinhardt, R., McCombie, W. R., de la, B. M., Dedhia,
N., Blocker, H., Hornischer, K., Nordsiek, G., Agarwala, R.,
Aravind, L., Bailey, J. A., Bateman, A., Batzoglou, S., Birney,
E., Bork, P., Brown, D. G., Burge, C. B., Cerutti, L., Chen, H.
C., Church, D., Clamp, M., Copley, R. R., Doerks, T., Eddy,
S. R., Eichler, E. E., Furey, T. S., Galagan, J., Gilbert, J. G.,
Harmon, C., Hayashizaki, Y., Haussler, D., Hermjakob, H.,
Hokamp, K., Jang, W., Johnson, L. S., Jones, T. A., Kasif, S.,
Kaspryzk, A., Kennedy, S., Kent, W. J., Kitts, P., Koonin, E.
V., Korf, I., Kulp, D., Lancet, D., Lowe, T. M., McLysaght, A.,
Mikkelsen, T., Moran, J. V., Mulder, N., Pollara, V. J., Ponting,
C. P., Schuler, G., Schultz, J., Slater, G., Smit, A. F., Stupka, E.,
Szustakowski, J., Thierry-Mieg, D., Thierry-Mieg, J., Wagner,
L., Wallis, J., Wheeler, R., Williams, A., Wolf, Y. I., Wolfe, K.
H., Yang, S. P., Yeh, R. F., Collins, F., Guyer, M. S., Peterson,
J., Felsenfeld, A., Wetterstrand, K. A., Patrinos, A., Morgan,
M. J., Szustakowki, J., de Jong, P., Catanese, J. J., Osoegawa,
K., Shizuya, H., & Choi, S. (2001). Initial sequencing and
analysis of the human genome. Nature, 409, 860–921.
Leblanc, R., Melanson, D., Tampieri, D., & Guttmann, R. D.
(1995). Familial cerebral aneurysms: A study of 13 families.
Neurosurgery, 37, 633–638.
 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
Leung, C. H., Poon, W. S., Yu, L. M., Wong, G. K., & Ng, H. K.
(2002). Apolipoprotein E genotype and outcome in aneurysmal
subarachnoid hemorrhage. Stroke, 33, 548–552.
Li, D. Y., Faury, G., Taylor, D. G., Davis, E. C., Boyle, W. A.,
Mecham, R. P., Stenzel, P., Boak, B., & Keating, M. T. (1998).
Novel arterial pathology in mice and humans hemizygous for
elastin. Journal of Clinical Investigation, 102, 1783–1787.
Liu, X., Wu, H., Byrne, M., Krane, S., & Jaenisch, R. (1997).
Type III collagen is crucial for collagen I fibrillogenesis and
for normal cardiovascular development. Proceedings of the
National Academy of Sciences of the United States of America,
94, 1852–1856.
Lonnqvist, L., Child, A., Kainulainen, K., Davidson, R., Puhakka,
L., & Peltonen, L. (1994). A novel mutation of the fibrillin
gene causing ectopia lentis. Genomics, 19, 573–576.
Lozano, A. M., & Leblanc, R. (1987). Familial intracranial
aneurysms. Journal of Neurosurgery, 66, 522–528.
Marquardt, G., Niebauer, T., Schick, U., & Lorenz, R. (2000).
Long term follow up after perimesencephalic subarachnoid
haemorrhage. Journal of Neurology, Neurosurgery and
Psychiatry, 69, 127–130.
Mirza, F. H., Smith, P. L., & Lim, W. N. (1979). Multiple
aneurysms in a patient with Ehlers-Danlos syndrome:
Angiography without sequelae. AJR American Journal of
Roentgenology, 132, 993–995.
Miskolczi, L., Guterman, L. R., Flaherty, J. D., & Hopkins, L. N.
(1998). Saccular aneurysm induction by elastase digestion of the
arterial wall: A new animal model. Neurosurgery, 43, 595–600.
Nienaber, C. A., & Von Kodolitsch, Y. (1999). Therapeutic
management of patients with Marfan syndrome: Focus on
cardiovascular involvement. Cardiology Reviews, 7, 332–341.
Niskakangas, T., Ohman, J., Niemela, M., Ilveskoski, E., Kunnas,
T. A., & Karhunen, P. J. (2001). Association of apolipoprotein
E polymorphism with outcome after aneurysmal subarachnoid
hemorrhage: A preliminary study. Stroke, 32, 1181–1184.
Onda, H., Kasuya, H., Yoneyama, T., Takakura, K., Hori,
T., Takeda, J., Nakajima, T., & Inoue, I. (2001).
Genomewide-linkage and haplotype-association studies map
intracranial aneurysm to chromosome 7q11. American Journal
of Human Genetics, 69, 804–819.
Ostergaard, J. R., & Oxlund, H. (1987). Collagen type III
deficiency in patients with rupture of intracranial saccular
aneurysms. Journal of Neurosurgery, 67, 690–696.
Peral, B., Ong, A. C., San Millan, J. L., Gamble, V., Rees, L., &
Harris, P. C. (1996). A stable, nonsense mutation associated
with a case of infantile onset polycystic kidney disease 1
(PKD1). Human Molecular Genetics, 5, 539–542.
Peters, D. G., Kassam, A. B., Feingold, E., Heidrich-O’Hare, E.,
Yonas, H., Ferrell, R. E., & Brufsky, A. (2001). Molecular
anatomy of an intracranial aneurysm: Coordinated expression
of genes involved in wound healing and tissue remodeling.
Stroke, 32, 1036–1042.
Peters, D. G., Kassam, A., St. Jean, P. L., Yonas, H., &
Ferrell, R. E. (1999). Functional polymorphism in the matrix
metalloproteinase-9 promoter as a potential risk factor for
intracranial aneurysm. Stroke, 30, 2612–2616.
1359
Pope, F. M., Nicholls, A. C., Narcisi, P., Bartlett, J., Neil-Dwyer,
G., & Doshi, B. (1981). Some patients with cerebral aneurysms
are deficient in type III collagen. Lancet, 1, 973–975.
Pritchard, C., Foulkes, L., Lang, D. A., & Neil-Dwyer, G. (2001).
Psychosocial outcomes for patients and carers after aneurysmal
subarachnoid haemorrhage. British Journal of Neurosurgery,
15, 456–463.
Putnam, E. A., Cho, M., Zinn, A. B., Towbin, J. A., Byers, P. H., &
Milewicz, D. M. (1996). Delineation of the Marfan phenotype
associated with mutations in exons 23–32 of the FBN1 gene.
American Journal of Medical Genetics, 62, 233–242.
Pyeritz, R. E. (2000). The Marfan syndrome. Annual Reviews in
Medicine, 51, 481–510.
Raaymakers, T. W. (1999). Aneurysms in relatives of patients
with subarachnoid hemorrhage: Frequency and risk factors.
MARS Study Group. Magnetic resonance angiography in
relatives of patients with subarachnoid hemorrhage. Neurology,
53, 982–988.
Reeders, S. T. (1992). Multilocus polycystic disease. Nature
Genetics, 1, 235–237.
Risch, N., & Merikangas, K. (1996). The future of genetic studies
of complex human diseases. Science, 273, 1516–1517.
Roberts, G. A., Corcoran, B. T., Pfouts, L. L., Phillips, J. P.,
Farrell, M. A., Bouchier-Hayes, D. J., & Collins, P. B. (2001).
Genetic evaluation of lipoprotein(a) in intracranial aneurysm
disease. Neurosurgery, 49, 133–140.
Robinson, P. N., Booms, P., Katzke, S., Ladewig, M., Neumann,
L., Palz, M., Pregla, R., Tiecke, F., & Rosenberg, T. (2002).
Mutations of FBN1 and genotype–phenotype correlations
in Marfan syndrome and related fibrillinopathies. Human
Mutation, 20, 153–161.
Ronkainen, A., Hernesniemi, J., & Tromp, G. (1995). Special
features of familial intracranial aneurysms: Report of 215
familial aneurysms. Neurosurgery, 37, 43–46.
Ronkainen, A., Miettinen, H., Karkola, K., Papinaho, S., Vanninen,
R., Puranen, M., & Hernesniemi, J. (1998). Risk of harboring
an unruptured intracranial aneurysm. Stroke, 29, 359–362.
Ronkainen, A., Puranen, M. I., Hernesniemi, J. A., Vanninen, R.
L., Partanen, P. L., Saari, J. T., Vainio, P. A., & Ryynanen, M.
(1995). Intracranial aneurysms: MR angiographic screening
in 400 asymptomatic individuals with increased familial risk.
Radiology, 195, 35–40.
Rosenorn, J., Eskesen, V., Schmidt, K., & Ronde, F. (1987).
The risk of rebleeding from ruptured intracranial aneurysms.
Journal of Neurosurgery, 67, 329–332.
Ryba, M., Grieb, P., Podobinska, I., Iwanska, K., Pastuszko, M., &
Gorski, A. (1992). HLA antigens and intracranial aneurysms.
Acta Neurochirurgica (Wien), 116, 1–5.
Saunders, A. M. (2000). Apolipoprotein E and Alzheimer disease:
An update on genetic and functional analyses. Journal of
Neuropathology and Experimental Neurology, 59, 751–758.
Schievink, W. I., Katzmann, J. A., Piepgras, D. G., & Schaid,
D. J. (1996). Alpha-1-antitrypsin phenotypes among patients
with intracranial aneurysms. Journal of Neurosurgery, 84(5),
781–784.
Schievink, W. I., Schaid, D. J., Michels, V. V., & Piepgras, D.
G. (1995). Familial aneurysmal subarachnoid hemorrhage: A
community-based study. Journal of Neurosurgery, 83, 426–429.
1360 B. Zhang et al. / The International Journal of Biochemistry & Cell Biology 35 (2003) 1341–1360
Schievink, W. I., Schaid, D. J., Rogers, H. M., Piepgras, D. G.,
& Michels, V. V. (1994). On the inheritance of intracranial
aneurysms. Stroke, 25, 2028–2037.
Sheffield, E. A., & Weller, R. O. (1980). Age changes at cerebral
artery bifurcations and the pathogenesis of berry aneurysms.
Journal of Neurological Sciences, 46, 341–352.
St. Jean, P., Hart, B., Webster, M., Steed, D., Adamson, J.,
Powell, J., & Ferrell, R. (1996). ␣1-Antitrypsin deficiency in
aneurysmal disease. Human Heredity, 46, 92–97.
Takenaka, K., Yamakawa, H., Sakai, H., Yoshimura, S., Murase,
S., Okumura, A., Nakatani, K., Kimura, T., Nishimura, Y.,
Yoshimi, N., & Sakai, N. (1998). Angiotensin I-converting
enzyme gene polymorphism in intracranial saccular aneurysm
individuals. Neurological Research, 20, 607–611.
Tartara, F., Gaetani, P., Tancioni, F., Guagliano, A., Klersy, C.,
Forlino, A., Marzatico, F., & Baena, R. (1996). ␣1-Antitrypsin
activity in subarachnoid hemorrhage. Life Science, 59, 15–20.
Tomasello, F., D’Avella, D., Salpietro, F. M., & Longo, M.
(1998). Asymptomatic aneurysms. Literature meta-analysis and
indications for treatment. Journal of Neurosurgical Sciences,
42(1 Supplement 1), 47–51.
Undlien, D. E., Lie, B. A., & Thorsby, E. (2001). HLA complex
genes in type 1 diabetes and other autoimmune diseases.
Which genes are involved? Trends in Genetics, 17, 93–100.
Vaicys, C., Hunt, C. D., & Heary, R. F. (2000). Ruptured
intracranial aneurysm in an adolescent with Alport’s
syndrome—a new expression of type IV collagenopathy: Case
report. Surgical Neurology, 54, 68–72.
van den Berg, J. S., Limburg, M., Kappelle, L. J., Pals, G.,
Arwert, F., & Westerveld, A. (1998). The role of type III
collagen in spontaneous cervical arterial dissections. Annals of
Neurology, 43, 494–498.
van den Berg, J. S., Pals, G., Arwert, F., Hennekam, R. C.,
Albrecht, K. W., Westerveld, A., & Limburg, M. (1999). Type
III collagen deficiency in saccular intracranial aneurysms.
Defect in gene regulation? Stroke, 30, 1628–1631.
van der, R. M., & Garrone, R. (1991). Collagen family of proteins.
The FASEB Journal, 5, 2814–2823.
Vaughan, C. J., Casey, M., He, J., Veugelers, M., Henderson,
K., Guo, D., Campagna, R., Roman, M. J., Milewicz, D. M.,
Devereux, R. B., & Basson, C. T. (2001). Identification of a
chromosome 11q23.2–q24 locus for familial aortic aneurysm
disease, a genetically heterogeneous disorder. Circulation, 103,
2469–2475.
Vermeulen, M., & van Gijn, J. (1990). The diagnosis of
subarachnoid haemorrhage. Journal of Neurology, Neurosurgery
and Psychiatry, 53, 365–372.
Wang, P. S., Longstreth Jr., W. T., & Koepsell, T. D.
(1995). Subarachnoid hemorrhage and family history. A
population-based case–control study. Archives of Neurology,
52, 202–204.
Weller, R. O. (1995). Subarachnoid haemorrhage and myths
about saccular aneurysms. Journal of Clinical Pathology, 48,
1078–1081.
Zhang, H., Apfelroth, S. D., Hu, W., Davis, E. C., Sanguineti, C.,
Bonadio, J., Mecham, R. P., & Ramirez, F. (1994). Structure
and expression of fibrillin-2, a novel microfibrillar component
preferentially located in elastic matrices. The Journal of Cell
Biology, 124, 855–863.
Zhang, B., Dhillon, S., Geary, I., Howell, W. M., Iannotti,
F., Day, I. N., & Ye, S. (2001). Polymorphisms in matrix
metalloproteinase-1, -3, -9, and -12 genes in relation to
subarachnoid hemorrhage. Stroke, 32, 2198–2202.
Zhang, M. C., He, L., Giro, M., Yong, S. L., Tiller, G. E., &
Davidson, J. M. (1999). Cutis laxa arising from frameshift
mutations in exon 30 of the elastin gene (ELN). The Journal
of Biological Chemistry, 274, 981–986.
Zhang, H., Hu, W., & Ramirez, F. (1995). Developmental
expression of fibrillin genes suggests heterogeneity of
extracellular microfibrils. The Journal of Cell Biology, 129,
1165–1176.