Alport Retinopathy Results from “Severe” COL4A5Mutations and Predicts Early Renal Failure

Rachel Tan, Deb Colville, Yan Yan Wang, Lin Rigby, and Judy SavigeUniversity of Melbourne Department of Medicine, Northern Health, Epping, Victoria, Australia

Background and objectives: Previous studies of X-linked Alport syndrome demonstrated that “severe” COL4A5 mutations(large deletions and rearrangements, nonsense and frame-shift mutations, and glycine substitutions in the carboxy-terminalresidues) were associated with early-onset renal failure, hearing loss, and lenticonus in affected male patients. This studyexamined whether severe mutations also resulted in the typical perimacular dot-and-fleck retinopathy.

Design, setting, participants, & measurements: Twenty unrelated families with X-linked Alport syndrome were studied forthe causative mutation in the COL4A5 gene. Nineteen affected male and 22 affected female individuals aged at least 14 yr fromthese families were examined for clinical and, in particular, ophthalmologic features.

Results: Nineteen pathogenic mutations were identified in the COL4A5 gene in the 20 families using a thermal melt analyzer(HRM RotorGene 6000; Corbett) or direct sequencing of hair root or skin fibroblast cDNA. Fifteen mutations were classifiedsevere and four as moderate. Severe mutations were associated with the central dot-and-fleck Alport retinopathy in maleindividuals (P � 0.0256) in addition to early-onset renal failure, hearing loss, and lenticonus (P � 0.0009, 0.009, and 0.009,respectively). Severe mutations did not correlate with clinical features in female individuals.

Conclusions: Severe mutations in male individuals with X-linked Alport syndrome are associated with the perimaculardot-and-fleck retinopathy. Furthermore, the retinopathy indicates that male individuals are at increased risk for renal failurebefore the age of 30 (P � 0.0007).

Clin J Am Soc Nephrol 5: 34–38, 2010. doi: 10.2215/CJN.01030209

A lport syndrome is an inherited kidney disease char-acterized by hematuria, progressive renal failure,hearing loss, lenticonus, and a dot-and-fleck retinop-

athy (1). It affects between one in 5000 and one in 50,000individuals (2) but is often unrecognized.

Approximately 85% of families with Alport syndrome haveX-linked disease (OMIM 301050) (3), which is due to mutationsin the COL4A5 gene (4). Male individuals are affected moreseverely than females, and hematuria is noted in early child-hood, but many males develop renal failure before the age of 30(“early-onset disease”). Most have a hearing loss, approxi-mately 40% develop lenticonus, and 60% have retinopathy(5,6). The hearing loss is helped with hearing aids, the lentico-nus eventually requires lens replacement, but the retinopathydoes not affect vision. The retinopathy is nevertheless helpfuldiagnostically because it is specific for Alport syndrome and isusually obvious especially in retinal photographs. Affected fe-male individuals with X-linked Alport syndrome typically havehematuria, and the likelihood of complications increases withincreasing age, with approximately 30% eventually developingrenal failure, hearing loss, or the ocular abnormalities (7,8).Clinical features in males are often consistent within families,

reflecting the strong genetic influence (5). In females, the clin-ical features are more variable in families because of lyoniza-tion (7).

Not only is identification of the underlying COL4A5 muta-tions diagnostic for X-linked Alport syndrome, but also thelocation and nature of the mutations help to predict the likeli-hood of clinical complications. Thus, males with large deletionsand nonsense and frame-shift mutations are more likely todevelop renal failure before the age of 30 yr, as well as hearingloss and lenticonus (5). In addition, missense mutations close tothe carboxy terminus result in a severe phenotype (6), butprevious studies have not correlated severe COL4A5 mutationswith an increased likelihood of retinopathy. The aim of thisstudy was to identify mutations in our patients with X-linkedAlport syndrome, confirm the previously proposed prognosticclassification of COL4A5 mutations, extend the classification toinclude the dot-and-fleck retinopathy, and determine whetherdemonstration of the retinopathy itself had any clinical signif-icance for predicting disease in other organs.

Materials and MethodsPatients

Twenty unrelated families with X-linked Alport syndrome werestudied. Alport syndrome was diagnosed on renal biopsy in 18 (90%)families, and X-linked inheritance was confirmed with linkage studiesin 15 (75%) (8). All participants were older than 14 yr, which is whenocular abnormalities are often first evident in male individuals.

Clinical features, including the presence and age at onset of end-stagerenal failure, and any clinically detectable hearing loss were noted.

Received February 11, 2009. Accepted September 29, 2009.

Published online ahead of print. Publication date available at www.cjasn.org.

Correspondence: Prof. Judy Savige, University of Melbourne, Northern Health,Northern Hospital, Epping VIC 3076, Australia. Phone: 03-8405-8823; Fax: 03-8405-8724; E-mail: jasavige@unimelb.edu.au

Copyright © 2010 by the American Society of Nephrology ISSN: 1555-9041/504–0034

Participants were then examined by an ophthalmologist who was notaware of their disease status. They were examined for the “oil-droplet”sign of anterior lenticonus, their pupils were dilated with 1% tropic-amide, and their optic fundi were examined for the perimacular dot-and-fleck retinopathy (Figure 1).

Mutation Detection MethodsTwo mutation detection methods were used. These were a genomic

DNA screening assay where variants were confirmed by direct se-quencing and an assay where hair root or skin fibroblast cDNA wasdirectly sequenced.

DNA Amplification and Screening. DNA was prepared fromperipheral venous blood using conventional methods. All 51 exons ofCOL4A5 plus flanking regions were amplified using primers based onpublished sequences (9) and screened for mutations using meltingcurve analysis (RotorGene 6000; Corbett). Amplicons with abnormalmelt curves were then sequenced.

Sequencing of Hair Root or Skin Fibroblast COL4A5 cDNA.mRNA was prepared from hair roots and immortalized fibroblast celllines (10–12) using Trizol (Invitrogen, Carlsbad, CA) following themanufacturer’s instructions. mRNA was then transcribed using theSuperscript III First-Strand Synthesis System (Invitrogen) using oli-godT as recommended by the manufacturer. Hair root cDNA wasamplified in nine overlapping fragments with primers designed usingPrimer3 design software (http://frodo.wi.mit.edu/primer3/). cDNAwere then amplified using PCR and directly sequenced for mutations.

SequencingAmplicons were purified using the UltraClean GelSpin DNA purifi-

cation kit (MoBio Laboratories, Carlsbad, CA) and sequenced using BigDye Terminator (Applied Biosystems, Foster City, CA) and an ABIPRISM 3100 Genetic Analyser (Applied Biosystems).

Definition of MutationsVariants were confirmed by sequencing in both directions. “Muta-

tions” were variants that segregated with disease within families andwere not present in 100 normal chromosomes. “Severe” mutations werelarge deletions and rearrangements, nonsense and frame-shift muta-tions, splicing mutations that affected the donor site, and missense

mutations in exons 23 through 51. All other mutations were classifiedas “moderate.”

Correlation of Mutations with Clinical FeaturesSevere and moderate mutations were correlated with clinical features,

namely the presence and age at onset of end-stage renal failure, and withthe presence of hearing loss, lenticonus, and the perimacular retinopathy.Results were compared with Fisher exact test (two-tailed; http://www.graphpad.com/quickcalcs/contingency2.cfm). This study was approvedby the Human Research Ethics Committee of Northern Health, and allparticipants provided signed informed consent.

ResultsMutations

Nineteen different mutations were identified in the 20 families(Table 1). G624D was present in two families who were not knownto be related but were both of Italian origin. Fourteen mutationswere novel. These included single-nucleotide substitutions, twoin-frame deletions, a frame-shift insertion, and a large rearrange-ment. Eighteen mutations affected the coding region, and oneaffected an intronic sequence. Coding region mutations werenonsense (n � 6) or missense (n � 8), and six of the missensemutations affected a glycine. Fifteen (71%) of 19 mutationswere classified severe and four as moderate. The mutations ofmoderate severity were c.142G�A, c.351_359delACCTCAAGC,c. 1165–1432del, and c.1498G�C.

Clinical FeaturesForty-one affected individuals were studied from the 20 fam-

ilies. These included 19 males and 22 females. Five families hadno affected adult male. Affected males had a median age of 38yr (range 14 to 68). Nine (47%) of 19 had reached end-stagerenal failure before the age of 30 yr, 15 (79%) of 19 had aclinically reported hearing loss, nine (53%) of 17 had lenticonus,and eight (50%) of 16 had a central retinopathy. The femaleindividuals had a median age of 41 yr (range 14 to 80). One (5%)of 22 had developed end-stage kidney failure before 30 yr, three(14%) of 22 had a hearing loss, none had lenticonus, and three(14%) of 21 had a central retinopathy.

Correlation of Mutations with Clinical FeaturesMale individuals with severe mutations were more likely to

develop early-onset renal failure, hearing loss, lenticonus, andretinopathy than those with moderate mutations (Table 2). Thedemonstration of the central retinopathy in males correlatedwith early-onset renal failure (P � 0.0070). Too few femalesdeveloped renal failure or other clinical features to determineany effect of mutation severity; however all four female indi-viduals who developed renal failure or the central retinopathyhad severe mutations.

DiscussionThis study identified 19 different COL4A5 mutations in 20

families with X-linked Alport syndrome by using a meltingcurve mutation screening assay or by direct sequencing of thehair root transcript. Fourteen mutations were novel, and 15were predicted to result in severe disease. Individuals withsevere mutations were more likely to present for medical at-

Figure 1. Alport dot-and-fleck retinopathy with perimaculardots and flecks.

Clin J Am Soc Nephrol 5: 34–38, 2010 Alport Retina and Renal Failure 35

Tab

le1.

CO

L4A

5m

utat

ions

det

ecte

din

X-l

inke

dA

lpor

tsy

ndro

me

Exo

nN

ucle

otid

ePr

otei

nC

onse

rved

Am

ino

Aci

dT

ype

ofC

hang

eC

olla

geno

usD

omai

nA

min

oA

cid

Cha

nge

Seve

re

2c.

87C

�A

ap.

C29

XY

esN

onse

nse

Yes

Neu

tral

tost

opY

es3

c.14

2G�

Aa

p.G

48R

Yes

Mis

sens

eY

esN

eutr

alto

neut

ral

No

6c.

351_

359d

elA

CC

TC

AA

GG

aN

AN

AIn

-fra

me

del

etio

nY

esN

AN

o19

c.11

17C

�T

p.R

373X

Yes

Non

sens

eY

esB

asic

tost

opY

es20

c.12

11_1

212i

nsA

aN

AN

AFr

ame-

shif

tin

sert

ion

Yes

NA

Yes

20/

21c.

1165

–143

2del

ap.

G38

9_K

474d

elN

AIn

-fra

me

del

etio

nY

esN

AN

o22

c.14

98G

�C

ap.

G50

0RY

esM

isse

nse

Yes

Neu

tral

toba

sic

No

24c.

1690

G�

Ta

p.G

564C

Yes

Mis

sens

eY

esN

eutr

alto

neut

ral

Yes

25c.

1871

G�

Ap.

G62

4DY

esM

isse

nse

Yes

Neu

tral

toac

idic

Yes

27IV

S26–

18A

�G

aN

AN

AIn

tron

icva

rian

tY

esN

AY

es30

c.24

73G

�T

ap.

G82

5XY

esN

onse

nse

Yes

Neu

tral

tone

utra

lY

es32

c.27

22G

�A

ap.

G90

8RY

esM

isse

nse

Yes

Neu

tral

toba

sic

Yes

33c.

2858

G�

Tp.

G95

3VY

esM

isse

nse

Yes

Alip

hati

c(n

och

ange

)Y

es43

c.39

25-?

del

ap.

G13

09_?

del

NA

Lar

gere

arra

ngem

ent

Yes

NA

Yes

44c.

4006

G�

Ta

p.G

1336

XY

esN

onse

nse

Yes

Neu

tral

tost

opY

es45

c.41

47C

�T

ap.

Q13

83X

Yes

Non

sens

eY

esA

cid

tost

opY

es48

c.45

49C

�A

p.P1

517T

Yes

Mis

sens

eN

oN

eutr

alto

neut

ral

Yes

49c.

4702

G�

Ca

p.E

1568

QY

esM

isse

nse

No

Neu

tral

tone

utra

lY

es51

c.50

29C

�T

p.R

1677

XY

esN

onse

nse

No

Neu

tral

tone

utra

lY

es

NA

,not

appl

icab

le.a

Nov

elm

utat

ion.

36 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol 5: 34–38, 2010

tention because they had renal failure or their clinical featureswere characteristic of Alport syndrome.

Two previous studies defined severe mutations in X-linkedAlport syndrome and demonstrated that they correlated withearly-onset renal failure, hearing loss, and lenticonus in maleindividuals; however, both studies failed to demonstrate a re-lationship with the perimacular dot-and-fleck retinopathy (5,6),possibly because their patients included children and becausepatients were examined in multiple centers by different oph-thalmologists who were unfamiliar with variants of the Alportretinopathy (13). Although our study was small, it did notinclude children who were younger than 14 yr, which is whenthe retinopathy is often first obvious in male individuals, andall but three participants were examined by the same ophthal-mologist. In many cases, retinal photographs were also exam-ined. This study not only confirmed the previously describedprognostic classification of COL4A5 mutations in male individ-uals with X-linked disease and that severe mutations correlatedwith early-onset renal failure, hearing loss, and lenticonus butalso demonstrated that severe mutations correlated with theperimacular dot-and-fleck retinopathy. Furthermore, this studyshowed that the retinopathy in male individuals with X-linkedAlport syndrome correlated with an increased likelihood ofearly-onset renal failure.

This study also demonstrated early-onset renal failure, lenti-conus, and retinopathy were uncommon in affected females,but females with these features all had severe mutations. Thishas implications for the increased risk for early-onset renalfailure in their male relatives but needs to be confirmed in alarger series. The development of renal failure and other clin-ical features in female individuals themselves is complicated bylyonization.

Not only does the demonstration of COL4A5 mutations inpatients with suspected Alport syndrome confirm the diagnosisof X-linked disease, but also the location of the mutation and itsnature (nonsense versus missense, frame-shift versus in-frameindel) helps to predict the clinical course in male individualsand, in particular, the likelihood of early-onset renal failure,

lenticonus, and, as demonstrated here, perimacular retinopa-thy; however, mutation detection may not always be necessary,because the perimacular dot-and-fleck retinopathy both con-firms the diagnosis of Alport syndrome and may predict in-creased risk for early-onset renal failure and hence the need forearly and aggressive management.

AcknowledgmentsThis work was supported by the National Health and Medical Re-

search Council of Australia; Kidney Health Australia; and the NorthernHealth Education, Equipment and Research Fund.

Preliminary results from this work were presented in abstract form atthe annual meeting of the American Society of Nephrology; November2 through 5, 2007.

We thank the many patients and their families who participated inthese studies.

DisclosuresNone.

References1. Gubler M, Levy M, Broyer M, Naizot C, Gonzales G, Perrin

D, Habib R: Alport’s syndrome: A report of 58 cases and areview of the literature. Am J Med 70: 493–505, 2001

2. Levy M, Feingold J: Estimating prevalence in single-genekidney diseases progressing to renal failure. Kidney Int 58:925–943, 2000

3. Feingold J, Bois E, Champret A, Broyer M, Gubler M-C,Grunfeld J-P: Genetic heterogeneity of Alport syndrome.Kidney Int 27: 672–677, 1985

4. Barker DF, Hostikka SL, Zhou J, Chow LT, Oliphant AR,Gerken SC, Gregory MC, Skolnick MH, Atkin CL, Tryg-gvason K: Identification of mutations in the COL4A5 genein Alport syndrome. Science 24: 1224–1227, 1990

5. Jais JP, Knebelmann B, Giatras I, De Marchi M, Rizzoni G,Renieri A, Weber M, Gross O, Netzer KO, Flinter F, PirsonY, Verellen C, Wieslander J, Persson U, Tryggvason K,Martin P, Hertz JM, Schroder C, Sanak M, Krejcova S,Carvalho MF, Saus J, Antignac C, Smeets H, Gubler MC:

Table 2. Frequency of clinical features associated with severe and moderate mutations

Parameter SevereMutations

ModerateMutations P

Males n � 12 n � 7renal failure before age 30a 9/10 0/6 0.0009clinically reported hearing loss 12/12 3/7 0.009lenticonus 9/12 0/5 0.009central retinopathyb 8/11 0/5 0.0256

Females n � 15 n � 7renal failure before age 30a 1/15 0/7 1.000clinically reported hearing loss 2/15 1/7 1.000lenticonus 0/15 0/7 1.000central retinopathyb 3/15 0/6 0.5263

aOf those who had already reached the age of 30.bOne adult had corneal scarring that precluded retinal examination, and three others did not have an ophthalmic

examination.

Clin J Am Soc Nephrol 5: 34–38, 2010 Alport Retina and Renal Failure 37

X-linked Alport syndrome: Natural history in 195 familiesand genotype–phenotype correlations in males. J Am SocNephrol 11: 649–657, 2000

6. Gross O, Netzer K-O, Lambrecht R, Seibold S, Weber M:Meta-analysis of genotype–phenotype correlation inX-linked Alport syndrome: Impact on clinical counselling.Nephrol Dial Transplant 17: 1218–1227, 2002

7. Jais JP, Knebelmann B, Giatras I, De Marchi M, Rizzoni G,Renieri A, Weber M, Gross O, Netzer KO, Flinter F, PirsonY, Dahan K, Wieslander J, Persson U, Tryggvason K, Mar-tin P, Hertz JM, Schroder C, Sanak M, Carvalho MF, SausJ, Antignac C, Smeets H, Gubler MC: X-linked Alportsyndrome: Natural history and genotype–phenotype cor-relations in girls and women belonging to 195 families—A“European Community Alport Syndrome Concerted Ac-tion” study. J Am Soc Nephrol 14: 2603–2610, 2003

8. Dagher H, Buzza M, Colville D, Jones C, Powell H, FassettR, Wilson D, Agar J, Savige J: A comparison of the clinical,histopathological and ultrastructural phenotypes in carri-ers of X-linked and autosomal recessive Alport syndrome.Am J Kidney Dis 38: 1217–1228, 2001

9. Martin P, Heiskari N, Zhou J, Leinonen A, Tumelius T,

Hertz JM, Barker D, Gregory M, Atkin C, Styrkarsdottir U,Neumann H, Springate J, Shows T, Petterson E, Tryggva-son K: High mutation detection rate in the COL4A5 colla-gen gene in suspected Alport syndrome using PCR anddirect DNA sequencing. J Am Soc Nephrol 9: 2291–2301,1998

10. King K, Flinter FA, Green PM: Hair roots as the idealsource of mRNA for genetic testing. J Med Genet 38: E20,2001

11. Tazon-Vega B, Ars E, Burset M, Santin S, Ruiz P, Fernan-dez-Llama P, Ballarin J, Torra R: Genetic testing for X-linked Alport syndrome by direct sequencing of COL4A5cDNA from hair root samples. Am J Kidney Dis 50: 257.e1–257.e14, 2007

12. Wang F, Wang Y, Ding J, Yang J: Detection of mutations inthe COL4A5 gene by analysing cDNA of skin fibroblasts.Kidney Int 67: 1268–1274, 2005

13. Colville D, Wang YY, Tan R, Savige J: The retinal “lozenge”or “dull macular reflex” in Alport syndrome is associatedwith a severe retinopathy and early onset renal failure. Br JOphthalmol 93: 383–386, 2009

38 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol 5: 34–38, 2010