Filaggrin Gene Defects Are Independent Risk Factors for Atopic Asthma in a Polish Population: A Study in ECAP Cohort
et al. (2011) Filaggrin Gene Defects Are Independent Risk Factors for Atopic
Asthma in a Polish Population: A Study in ECAP Cohort. PLoS ONE 6(2): e16933. doi:10.1371/journal.pone.0016933
Filaggrin Gene Defects Are Independent Risk Factors for Atopic Asthma in a Polish Population: A Study in ECAP Cohort
Joanna Ponin ska 0
Bolesaw Samolin ski 0
Aneta Tomaszewska 0
Filip Raciborski 0
Piotr Samel-Kowalik 0
Artur Walkiewicz 0
Agnieszka Lipiec 0
Barbara Piekarska 0
Jarosaw Komorowski 0
Edyta Krzych-Fata 0
Andrzej Namysowski 0
Jacek Borowicz 0
Graz_yna Kostrzewa 0
Sawomir Majewski 0
Rafa Poski 0
Jacques Zimmer, Centre de Recherche Public de la Sante (CRP-Sante), Luxembourg
0 1 Department of Medical Genetics, Medical University of Warsaw , Warsaw , Poland , 2 Department of Prevention of Environmental Hazards and Allergology, Medical University of Warsaw , Warsaw , Poland , 3 Department of Dermatology and Venereology, Medical University of Warsaw , Warsaw , Poland
Background: FLG null variants of which 2282del4 and R501X are the most frequent in Caucasians are established risk factors for atopic dermatitis (AD) with an effect probably mediated through impairment of epidermal barrier. Among subjects with AD FLG defects are also consistently associated with asthma and allergic rhinitis (AR) but it is less clear to what extent these associations are also present independently from skin disease. The aim of the present study was to evaluate the role of 2282del4 and R501X in predisposing to these allergic phenotypes in a Polish population. Methodology: 2282del4 and R501X were typed among 3,802 participants of the Epidemiology of Allergic Diseases in Poland (ECAP) survey, a cross-sectional population-based study using ECRHS II and ISAAC questionnaires, and ambulatory examination. Principal Findings: The FLG null variants were associated with AD (OR = 2.01, CI: 1.20-3.36, P = 0.007), allergic rhinitis (in particular persistent form, OR = 1.69, CI:1.12-2.54, P = 0.011), and asthma (in particular atopic asthma, OR = 2.22, CI:1.24-3.96, P = 0.006). Association with atopic asthma (but not persistent allergic rhinitis) was also present in the absence of AD, (OR = 2.02, CI: 1.07-3.81, P = 0.027) as well as in the absence of AD and history of broadly defined inflammatory skin disease (OR = 2.30, CI: 1.07-4.93, P = 0.03). Association to atopic asthma would have not been found if diagnosis was made by questionnaire only (OR = 1.15, CI: 0.58-2.32, P = 0.8). We did not observe an association between FLG variants and allergic sensitizations (P = 0.8) or total IgE. (P = 0.6). Conclusions/Significance: In a Polish population FLG 2282del4 and R501X carriage increases risk for development of AD and atopic asthma (also in the absence of AD or history thereof). This suggests that interventions aimed at restoring epidermal barrier may have a general role in asthma prophylaxis/treatment.
Funding: The work was financed by grants from Polish Ministry of Science and Higher Education, Ministry of Health grant nr 6P052005C/06572, and Warsaw
Medical University grant nr 1WY/NK1W/2009. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the
Competing Interests: The authors have declared that no competing interests exist.
Filaggrin gene (FLG) is strongly expressed in the granular cells of
the epidermis leading to production of a large precursor protein
profilaggrin. In the process of differentiation profilaggrin is
proteolytically cleaved into functional filaggrin peptides which bind
and collapse the keratin cytoskeleton and subsequently are degraded
into hydrophilic amino acids forming the natural moisturizing
factor. The N-terminal domain of profilaggrin is likely to have an
additional function as it specifically localizes to the nucleus. All these
processes are critical for creation of epidermal barrier with
appropriate mechanical and biochemical properties .
FLG null variants are strong risk factor for AD [2,3]. In
Caucasians two such variants are particularly common: 2282del4
and R501X with originally reported carrier rates in general
population of ,2 and 6%, respectively . Both variants result in
a complete loss of processed filaggrin due to premature
termination codons within the first FLG repeat. Whereas several
new FLG variants have been reported they are substantially less
prevalent and qualitatively different with some residual function
Among subjects with AD FLG defects are associated with other
allergic disease such as asthma and allergic rhinitis (AR) however,
it is less clear to what extent these associations are present
independently from skin disease . Two meta-analyses concluded
that there was no association between FLG null variants and
asthma among subjects without AD although the ORs from
pooled estimates suggested a trend in the direction of association
[5,6]. Regarding association with AR in the absence of AD two
studies reported conflicting results: Weideinger et al.  found an
effect whereas Marenholz et al. did not .
Our purpose was to examine in a Polish population association
of the 2282del4 and R501X FLG loss-of-function variants with
AD, asthma and allergic rhinitis.
Materials and Methods
The study was approved by Ethical Committee of Medical
University of Warsaw. All ECAP subjects gave an informed
written consent including specific consent to genetic testing.
Written consent for anonymous use of DNA was also obtained
from subjects undergoing paternity tests whose samples were used
to verify population prevalence of R501X.
The study was based on participants of Epidemiology of Allergic
Diseases in Poland (ECAP, www.ECAP.pl) living in major
metropolitan areas of Poland (Katowice, Wrocaw, Lublin,
Gdan sk, Warszawa, Poznan and Biaystok). ECAP is a
continuation of the European Community Respiratory Health Survey II
(ECRHS II) and International Study of Asthma and Allergy in
Childhood (ISAAC). ECAP includes randomly selected population
aged 2044 y.o. (ECRHS standard) as well as 67 y.o. and 1314
y.o. (ISAAC standard). The recruitment was done by a
randomization procedure based on the personal identity number
(PESEL). Out of the 25262 subjects who were approached by
pollsters, 9446 refused participation (response rate 262.6%). In
the present analysis two of the questionnaires answers were used
(i) Have you ever had asthma?, (ii) Have you ever had eczema,
atopic dermatitis or other inflammatory skin condition.
Those who completed questionnaire were invited for an
ambulatory examination. Four thousand thirty eight subjects
(25.5%) have taken up the offer. The examination included
medical history, physical examination, spirometry, PNIF (Peak
Nasal Inspiratory Flow) and skin prick tests with 15 allergens:
hazel, alder, birch, grasses/grain, rye, Artemisia, plantain,
Alternaria, Cladosporium, molds I (Alternaria tenuis, Botrytis
cinerea, Cladosporium herbarum, Culvularia lunata, Fusarium
moniliforme, Helminthosporium), molds II (Aspergillus fumigatus,
Mucor mucedo, Penicillium notatum, Rhizopus nigricans, Serpula
lacrymans, Pullularia pullulans), Dermatophagoides pteronyssinus,
Dermatophagoides farinae, dog, cat, negative control, histamine.
Concentration of total IgE in serum was determined with
reagents of the Phadia CAP System  (N = 3440) or the
Allergopharma-ELISA-Test  (N = 712). The obtained data
were presented in IU/ml (international units per mililitre).
The clinical diagnoses of asthma (atopic or non atopic),
intermittent allergic rhinitis (i.e. with symptoms present ,4 days
a week or for ,4 consecutive weeks), persistent allergic rhinitis,
(i.e. with symptoms present .4 days a week and for .4
consecutive weeks) and atopic dermatitis were based on the
International Global Initiative for Asthma (GINA) guidelines ,
ARIA criteria [12,13], and criteria of Hanifin and Rajka , for
asthma, allergic rhinitis and atopic dermatitis, respectively. In
addition a history of food allergy, drug allergy, insect bite allergy,
urticaria, Quinckes oedema or other chronic diseases was
obtained. During the examination blood samples were collected.
While analyzing associations between FLG variants and allergic
disorders comparisons were performed against a group of healthy
controls (N = 1865), defined as individuals without any allergic
disorder or other chronic disease (including ichtyosis vulgaris)
based on performed clinical workup and history. Family history of
atopy or other disease(s) was not an exclusion criterion.
Due to faulty blood sample collection (wrong labeling,
degradation) the final genetic analysis was carried in 94% of
those who underwent medical exam, i.e. 3802 subjects: 951
children 67 y.o. (47.6% females), 1054 children 1314 y.o.
(49.2% females) and 1797 adults (60.7% females).
Genomic DNA was extracted from whole blood. Typing for
2282del4 was performed by sizing of fluorescently labeled PCR
product on ABI 3130 sequencer, typing for R501X was done by
PCR-RFLP with NlaIII restrictase (1639 samples) or TaqMan
allelic discrimination assay (2323 samples). One hundred sixty
samples were typed for R501X by both methods with complete
concordance. All these methods were described previously
[2,8,15]. During typing every positive sample was repeated by
reanalysis of DNA from the original stock. Whole screening was
blinded to diagnoses.
In the analysis heterozygous and homozygous genotypes were
pooled. Statistical significance of differences in genotype frequency
among analyzed groups was assessed with chi square test or
Fishers exact test as appropriate. The strength of association was
estimated by calculating Odss Ratio (OR) with 95% confidence
interval (CI). Given the reports on the role of the FLG null variants
in predisposing to AD, asthma and AR no correction for multiple
testing was applied. Deviation from Hardy-Weinberg equilibrium
was assessed by a Chi-square test with one degree of freedom.
Total IgE concentration was log transformed to achieve normal
distribution and presented means are geometric means (i.e. back
transformed). In the analysis of FLG status vs. total IgE the
adjustment for test-method, sex and age was performed by
In our study we could detect the following effects with the power
of 0.8 at alpha = 0.05: asthmaOR = 2.0, allergic rhinitis
OR = 1.7, atopic dermatitisOR = 2.3 and allergic sensitization
OR = 1.5.
Calculations were performed with Statistica package.
The FLG R501X variant is rare in a Polish population
While analyzing the 3802 subjects from ECAP cohort we
identified 3629 wild type, 140 heterozygous and three homozygous
2282del4 genotypes (carriage rate: 3.76%, CI: 3.204.41) as well
as 30 heterozygous R501X genotypes (carriage rate: 0.8%, CI:
0.551.12). There were no compound heterozygotes. In order to
verify relatively low frequency of R501X vs. 2282del4 variant in a
Polish population we also tested 510 samples randomly selected
from an anonymous bank containing DNA isolated for the
purpose of paternity tests . We found 19 samples positive for
2282del4 (carriage rate: 3.79%, CI: 2.448.85) and five positive for
R501X (carriage rate: 1.0%, CI: 0.432.31). The distribution of
FLG variants did not deviate from Hardy-Weinberg equilibrium in
either cohort (P = 0.5 in both cases).
Associations between FLG variants and studied
When analyzing distribution of combined FLG variants we
found an association with AD (OR = 2.01, P = 0.007), asthma
(OR = 1.70, P = 0.024) and AR (OR = 1.43, P = 0.046, Table 1).
Analysis of individual variants showed statistically significant
associations between 2282del4 and AD (OR = 1.92, P = 0.022),
asthma (OR = 1.97, P = 0.005), and AR (OR = 1.47, P = 0.047,
Further analysis indicated that the statistically significant
associations between 2282del4 or combined genotype and asthma
as well as AR were limited, respectively, to atopic asthma (AA) and
persistent AR (pAR, Table 1).
We noted that when subjects were stratified according to the
answer to the question: Have you ever had asthma the frequency
of FLG variants was statistically significantly increased only among
those who were diagnosed with asthma by a physician during the
present study but who were not aware of having the disease as
judged by questionnaire data (OR = 1.81, P = 0.03 and OR =
3.53, P = 0.00005, for the comparison of combined genotype
frequency vs. healthy controls, for all asthma and AA, respectively,
Table 2). Had the study been based solely on questionnaire data,
only a trend for association between combined FLG variants and
asthma would have been found (OR = 1.15, P = 0.83, Table 2).
In ECAP cohort there was no association between FLG variants
and allergic sensitizations or total IgE concentration. The
prevalence of the combined genotype was 4.7% (78/1676) vs.
4.5% (95/2126) among those with a positive skin-prick test to at
least one allergen and the remaining group, respectively
(OR = 1.04, CI:0.771.42, P = 0.8). Mean concentration of total
IgE was 9.08 (CI: 7.3811.17) vs. 9.62 (CI: 9.2010.06) for those
with and without FLG defects, respectively (P = 0.6, analysis
adjusted for test-method, sex and age category).
The association between FLG null variants and atopic
asthma (AA) is also found among those without atopic
dermatitis (AD) or history thereof
Although the OR for AA conferred by 2282del4 or the
combined genotype was higher among those with than those
without AD (OR = 4.37 vs. OR = 2.23, and OR = 3.61 vs.
OR = 2.02, respectively) the associations were statistically
significant in both subgroups (P,0.03, Table 3). Analysis of pAR
showed similar trends although the statistical significance of these
associations among those without AD was borderline (P = 0.049
and P = 0.053, for 2282del4 and combined FLG variants,
respectively (Table 4).
The frequency of the combined genotype showed a trend for
increase among subjects who were not diagnosed with AD but
who reported history of an inflammatory skin condition in the
questionnaire (OR = 1.38, CI: 0.962.0, P = 0.08, comparison vs.
healthy controls). Thus, we were interested whether the observed
associations between FLG variants and AA among those without
AD could be caused by an association among those with a history
of AD or other inflammatory skin disease. However, this was not
apparent since there was an association between 2282del4 or
combined genotype and AA also among subjects without AD
according to both clinical diagnosis and self reported history of an
inflammatory skin condition: 7.1% (7/99) vs. 3.1% (55/1790),
OR = 2.40 (CI: 1.065.42), P = 0.03 and 8.1% (8/99) vs. 3.7%
(66/1790), OR = 2.30 (CI: 1.074.93), P = 0.03, for 2282del4 and
combined genotype, respectively.
Since the chances of AD resolution increase with age we also
analyzed the association between AA and FLG variants among
those without AD in the youngest age group (i.e. children 67 y.o.).
Among those with AA the prevalence of 2282del4 and the
combined genotype was 12.9% (4/31) which was higher than
among controls (OR = 4.53, CI: 1.5413.38, P = 0.018 and
OR = 3.74, CI: 1.2810.98, P = 0.032, for 2282del4 and the
combined genotype, respectively).
Conversely, analysis of pAR did not show associations with FLG
variants among those without AD according to both the
questionnaire and clinical diagnosis: 3.8% (13/340) vs. 3.1%
(55/1790), OR = 1.25 (CI: 0.682.32, P = 0.5), and 4.7% (16/340)
vs. 3.7% (66/1790), OR = 1.29 (CI: 0.742.26, P = 0.4) for
prevalence of 2282del4 and the combined genotype among pAR
and healthy controls, respectively.
While studying a population based cohort of subjects we
observed that the FLG defects conferred an increased risk for
development of AD, AR (in particular pAR) and asthma (in
particular AA). Whereas both associations were particularly strong
among subjects with AD, the association with AA remained after
exclusion of subjects with current AD even when analysis was
limited to the youngest age group, i.e. a group with the lowest
chance of complete resolution of skin disease. Association between
AA and FLG variants was also present among those without
current AD or history of AD or other inflammatory skin disease.
The association between FLG defects and AA in the absence of
AD contrasts with conclusions of two recent meta-analyses
2282del4 or R501X
1In 24 subjects allergic rhinitis could not be classified as intermittent or persistent; NA not applicable; All comparisons vs. healthy controls.
1.47 (0.683.22) 0.34
1.81 (1.073.07) 0.03
0.86 (0.233.25) 0.9
3.53 (1.866.72) 0.00005
1.15 (0.582.32) 0.83
Asthma (all) by questionnaire irrespective 206
of physicians diagnosis
2282del4 or R501X
* Calculated for the comparison of the frequency of the combined genotype (2282del4 or R501X) vs. healthy controls. Cells with P values ,0.05 are boldfaced;
Questionnaire data were not available in 18 subjects with asthma including 9 with AA.
although it should be noted that both these studies reported trends
in the direction of association (OR = 1.11 and OR = 1.30) [5,6].
On one hand, some cases of resolved AD might have been missed
in our study due to lack of patients/parents recall. Recall errors
regarding history of allergic diseases have been demonstrated 
and are likely to exist also in our cohort. On the other hand, the
discrepancy with pervious studies [5,6] might also be caused by
population specific genetic, environmental and/or life style factors
as well as methodological issues. In our cohort the association
between FLG null variants and asthma (in particular AA) was found
preferentially (exclusively?) among those who were not aware of
having the disease. This suggests that AA associated with FLG
defects in the absence of AD may have a subtle phenotype being
particularly difficult to diagnose by family practitioners.
Notwithstanding the precise reasons for the discussed discrepancies, our
results indicate that in a Polish population FLG defects represent a
risk factor for asthma, irrespective of apparent skin disease or history
thereof which it is possible to elicit in a clinical setting.
Interestingly, association between FLG defects and asthma
without eczema has also been found in a cohort of Danish children
prospectively followed from birth . Furthermore, similar
longitudinal follow-up methodology which should maximize the
diagnosis rate was also employed in a study of German cohort
where a relatively distinct trend towards an association was found
(OR = 2.47, P = 0.11) . Further evidence implicating epidermal
barrier function in asthma pathogenesis came recently from a
large genome-wide study showing that a locus with a likely
function in keratinocytes (RORA) was among ten loci most strongly
associated with this disease . These findings suggest that at
least in some cohorts epidermal barrier defects may play a role in
asthma pathogenesis among those without AD.
In contrast to studies in other populations [5,6] we did not
observe an association between FLG null variants and allergic
sensitization(s) as judged by analysis of skin prick test results or
concentration of total IgE. This result is consistent with recent
observations in a Danish cohort where the risk of sensitization
among FLG defect carriers increased only after onset of asthma
and/or eczema  and suggests that the effect of FLG null
variants on AD or AA development is not likely to be primarily
mediated through allergic sensitization.
The association between FLG variants and AD confirms the
findings in other populations [3,6]. However, the association
found in our study had only moderate statistical significance and
effect size. This is consistent with suggestions that FLG variants are
associated with severe forms of AD which are more readily
ascertained in hospital based studies .
Our results also add to data on differences in population specific
prevalence of FLG variants. We showed that in a Polish population
the prevalence of the R501X variant (,1%) was distinctly lower
than the prevalence of ,6% reported for Irish and Scottish
populations . An intermediate R501X frequency in German
population (,2.5% as estimated from pooled data of Stemmler
et al.  and Weidinger et al. ) suggests a clinal variation in
prevalence of this variant in Europe.
In conclusion, we show that in a Polish population FLG null
variants 2282del4 and R501X are risk factors for AD, and
independently from it, for AA. A methodological observation is
that in a Polish population AA associated with FLG defects may
All comparisons vs. healthy controls (Table 1), NA: not available.
Table 4. Distribution of the FLG variants in subjects with persistent allergic rhinitis (pAR) stratified by diagnosis of atopic
All comparisons vs. healthy controls (Table 1), NA: not available.
Kuna ( odz), Barbara Rogala and Radosaw Gawlik (Katowice) for help in
carrying out the study.
We would like to thank Andrzej Emeryk (Lublin), Ewa
Niz_ankowskaMogilnicka (Krakow), Andrzej Fal (Wrocaw), Wojciech Silny (Pozna n),
Anna Bodzenia-ukaszyk (Biaystok), Ewa Jassem (Gdan sk), Anna
Breborowicz (Pozna n), Jurek Kruszewski, Marek Kulus (Warszawa), Piotr
Conceived and designed the experiments: RP BS SM JP. Performed the
experiments: JP GK. Analyzed the data: RP JP BS. Contributed reagents/
materials/analysis tools: AT FR PS-K AW AL BP JK EK-F AN JB. Wrote
the paper: RP BS JP SM.
1. O'Regan GM , Sandilands A , McLean WHI , Irvine AD ( 2008 ) Filaggrin in atopic dermatitis . Journal of Allergy and Clinical Immunology 122 : 689 - 693 .
2. Palmer CNA , Irvine AD , Terron-Kwiatkowski A , Zhao Y , Liao H , et al. ( 2006 ) Common loss-of-function variants of the epidermal barrier protein filaggrin are a major predisposing factor for atopic dermatitis . Nat Genet 38 : 441 - 446 .
3. Baurecht Hr , Irvine AD , Novak N , Illig T , Buhler B , et al. ( 2007 ) Toward a major risk factor for atopic eczema: Meta-analysis of filaggrin polymorphism data . Journal of Allergy and Clinical Immunology 120 : 1406 - 1412 .
4. Sandilands A , Terron-Kwiatkowski A , Hull PR , O'Regan GM , Clayton TH , et al. ( 2007 ) Comprehensive analysis of the gene encoding filaggrin uncovers prevalent and rare mutations in ichthyosis vulgaris and atopic eczema . Nat Genet 39 : 650 - 654 .
5. van den Oord RAHM , Sheikh A ( 2009 ) Filaggrin gene defects and risk of developing allergic sensitisation and allergic disorders: systematic review and meta-analysis . BMJ 339: b2433.
6. Rodriguez E , Baurecht H , Herberich E , Wagenpfeil S , Brown SJ , et al. ( 2009 ) Meta-analysis of filaggrin polymorphisms in eczema and asthma: Robust risk factors in atopic disease . Journal of Allergy and Clinical Immunology 123 : 1361 - 1370 .
7. Weidinger S , O'Sullivan M , Illig T , Baurecht Hr , Depner M , et al. ( 2008 ) Filaggrin mutations, atopic eczema, hay fever, and asthma in children . Journal of Allergy and Clinical Immunology 121 : 1203 - 1209 .
8. Marenholz I , Nickel R , Ruschendorf F , Schulz F , Esparza-Gordillo J , et al. ( 2006 ) Filaggrin loss-of-function mutations predispose to phenotypes involved in the atopic march . Journal of Allergy and Clinical Immunology 118 : 866 - 871 .
9. Johansson GOS ( 1998 ) Clinical Workshop . IgE antibodies and the Pharmacia CAP System in allergy diagnosis . Lidkoping: Landstroms . 48 p.
10. Debelic M , Wahl R ( 1996 ) In vitro tests: immunoglobulins E and G . In: Fuchs E, Schulz KH, eds. Manuale allergologicum IV. Deisenhofen , Germany: DustriVerlag Dr Karl Feistle . pp 1 - 27 .
11. Global Initiative for Asthma ( 2009 ) Global Strategy for Asthma . Management and Prevention NHLBI/WHO Workshop Report . http://www.ginasthma.com.
12. Bousquet J , van Cauwenberge P , Khaltaev N ( 2001 ) Allergic rhinitis and its impact on asthma . J Allergy Clin Immunol 108 : S147 - S334 .
13. Bousquet J , Khaltaev N , Cruz AA , Denburg J , Fokkens WJ , et al. ( 2008 ) Allergic rhinitis and its impact on asthma (ARIA) 2008 update (in collaboration with the World Health Organization, GA(2)LEN and AllerGen) . Allergy 63 : 8 - 160 .
14. Hanifin JM , Rajka G ( 1980 ) Diagnostic features of atopic dermatitis . Acta Derm Venereol (supl) 92 : 44 - 47 .
15. Smith FJ , Irvine AD , Terron-Kwiatkowski A , Sandilands A , Campbell LE , et al. ( 2006 ) Loss-of-function mutations in the gene encoding filaggrin cause ichthyosis vulgaris . Nat Genet 38 : 337 - 342 .
16. Mueller-Malesinska M , Nowak M , Ploski R , Waligora J , Korniszewski L ( 2001 ) Epidemiology of 35delG mutation in GJB2 gene in a Polish population . Journal of Audiological Medicine 10 : 136 - 141 .
17. Kulig M , Bergmann R , Edenharter G , Wahn U ( 2000 ) Does allergy in parents depend on allergy in their children? Recall bias in parental questioning of atopic diseases . Multicenter Allergy Study Group. J Allergy Clin Immunol 105 : 274 - 278 .
18. Bonnelykke K , Pipper CB , Tavendale R , Palmer CN , Bisgaard H ( 2010 ) Filaggrin gene variants and atopic diseases in early childhood assessed longitudinally from birth . Pediatr Allergy Immunol 21 : 954 - 961 .
19. Moffatt MF , Gut IG , Demenais F , Strachan DP , Bouzigon E , et al. ( 2010 ) A Large-Scale, Consortium-Based Genomewide Association Study of Asthma . New England Journal of Medicine 363 : 1211 - 1221 .
20. Weidinger S , Rodriguez E , Stahl C , Wagenpfeil S , Klopp N , et al. ( 2006 ) Filaggrin Mutations Strongly Predispose to Early-Onset and Extrinsic Atopic Dermatitis . J Invest Dermatol 127 : 724 - 726 .
21. Brown SJ , Sandilands A , Zhao Y , Liao H , Relton CL , et al. ( 2007 ) Prevalent and Low-Frequency Null Mutations in the Filaggrin Gene Are Associated with Early-Onset and Persistent Atopic Eczema . J Invest Dermatol 128 : 1591 - 1594 .
22. Barker JNWN , Palmer CNA , Zhao Y , Liao H , Hull PR , et al. ( 2006 ) Null Mutations in the Filaggrin Gene (FLG) Determine Major Susceptibility to EarlyOnset Atopic Dermatitis that Persists into Adulthood . J Invest Dermatol 127 : 564 - 567 .
23. Henderson J , Northstone K , Lee SP , Liao H , Zhao Y , et al. ( 2008 ) The burden of disease associated with filaggrin mutations: A population-based, longitudinal birth cohort study . Journal of Allergy and Clinical Immunology 121 : 872 - 877 .
24. Stemmler S , Parwez Q , Petrasch-Parwez E , Epplen JT , Hoffjan S ( 2006 ) Two Common Loss-of-Function Mutations within the Filaggrin Gene Predispose for Early Onset of Atopic Dermatitis . J Invest Dermatol 127 : 722 - 724 .