Phenotypic Mutation 'aoba' (pdf version)
Alleleaoba
Mutation Type critical splice donor site (1 bp from exon)
Chromosome1
Coordinate82,513,461 bp (GRCm39)
Base Change C ⇒ T (forward strand)
Gene Col4a4
Gene Name collagen, type IV, alpha 4
Synonym(s) E130010M05Rik, [a]4(IV)
Chromosomal Location 82,426,144-82,564,570 bp (-) (GRCm39)
MGI Phenotype FUNCTION: [Summary is not available for the mouse gene. This summary is for the human ortholog.] This gene encodes one of the six subunits of type IV collagen, the major structural component of basement membranes. This particular collagen IV subunit, however, is only found in a subset of basement membranes. Like the other members of the type IV collagen gene family, this gene is organized in a head-to-head conformation with another type IV collagen gene so that each gene pair shares a common promoter. Mutations in this gene are associated with type II autosomal recessive Alport syndrome (hereditary glomerulonephropathy) and with familial benign hematuria (thin basement membrane disease). Two transcripts, differing only in their transcription start sites, have been identified for this gene and, as is common for collagen genes, multiple polyadenylation sites are found in the 3' UTR. [provided by RefSeq, Jul 2008]
PHENOTYPE: Mice homozygous for an ENU-induced mutation develop an early nephritic syndrome associated with uremia, proteinuria, hematuria, leukocyturia, and focal segmental glomerulosclerosis, and die prematurely of kidney failure. Some homozygotes exhibit moderatesensorineural hearing loss. [provided by MGI curators]
Accession Number

NCBI RefSeq: NM_007735; MGI: 104687

MappedYes 
Amino Acid Change
Institutional SourceBeutler Lab
Gene Model not available
AlphaFold Q9QZR9
SMART Domains Protein: ENSMUSP00000084282
Gene: ENSMUSG00000067158

DomainStartEndE-ValueType
low complexity region 29 41 N/A INTRINSIC
Pfam:Collagen 54 113 4e-11 PFAM
Pfam:Collagen 110 168 4.1e-10 PFAM
Pfam:Collagen 172 229 2.8e-10 PFAM
low complexity region 265 288 N/A INTRINSIC
internal_repeat_7 289 345 1.46e-9 PROSPERO
internal_repeat_6 291 348 5.03e-10 PROSPERO
internal_repeat_9 297 353 7.22e-9 PROSPERO
internal_repeat_4 322 354 2.06e-11 PROSPERO
internal_repeat_11 334 349 1.25e-5 PROSPERO
Pfam:Collagen 392 449 1.3e-8 PFAM
low complexity region 461 482 N/A INTRINSIC
Pfam:Collagen 486 553 1e-10 PFAM
low complexity region 563 595 N/A INTRINSIC
Pfam:Collagen 597 658 1e-8 PFAM
Pfam:Collagen 663 731 4.4e-10 PFAM
Pfam:Collagen 755 810 3.3e-9 PFAM
internal_repeat_2 816 841 2.9e-13 PROSPERO
Pfam:Collagen 844 912 1.8e-10 PFAM
Pfam:Collagen 898 962 2.7e-10 PFAM
low complexity region 963 1003 N/A INTRINSIC
Pfam:Collagen 1006 1071 2e-10 PFAM
Pfam:Collagen 1073 1132 5.8e-12 PFAM
Pfam:Collagen 1124 1185 1.8e-10 PFAM
Pfam:Collagen 1187 1245 2.3e-8 PFAM
low complexity region 1277 1361 N/A INTRINSIC
low complexity region 1371 1384 N/A INTRINSIC
Pfam:Collagen 1395 1454 4.3e-8 PFAM
C4 1457 1564 3.36e-58 SMART
C4 1565 1681 1.49e-59 SMART
Predicted Effect probably benign
Meta Mutation Damage Score Not available question?
Is this an essential gene? Probably nonessential (E-score: 0.089) question?
Phenotypic Category Autosomal Recessive
Candidate Explorer Status loading ...
Single pedigree
Linkage Analysis Data
Penetrance 100% 
Alleles Listed at MGI

All alleles(1) : Chemically induced(1)

Lab Alleles
AlleleSourceChrCoordTypePredicted EffectPPH Score
IGL00402:Col4a4 APN 1 82469362 missense unknown
IGL01092:Col4a4 APN 1 82444266 missense unknown
IGL01104:Col4a4 APN 1 82444266 missense unknown
IGL01413:Col4a4 APN 1 82448969 missense unknown
IGL01518:Col4a4 APN 1 82433480 missense unknown
IGL02014:Col4a4 APN 1 82501681 splice site probably benign
IGL02215:Col4a4 APN 1 82431530 missense unknown
IGL02707:Col4a4 APN 1 82471237 missense unknown
IGL02858:Col4a4 APN 1 82506204 missense unknown
IGL02987:Col4a4 APN 1 82476646 splice site probably benign
IGL03384:Col4a4 APN 1 82462159 missense probably benign 0.04
amazement UTSW 1 82458207 nonsense probably null
asombro UTSW 1 82466730 critical splice donor site probably null
astonishment UTSW 1 82433449 missense unknown
wonderment UTSW 1 82430865 missense unknown
IGL02980:Col4a4 UTSW 1 82447198 critical splice donor site probably null
R0028:Col4a4 UTSW 1 82465231 critical splice donor site probably null
R0083:Col4a4 UTSW 1 82484832 critical splice acceptor site probably null
R0696:Col4a4 UTSW 1 82470270 missense unknown
R0788:Col4a4 UTSW 1 82502717 missense unknown
R0789:Col4a4 UTSW 1 82502717 missense unknown
R0790:Col4a4 UTSW 1 82502717 missense unknown
R0894:Col4a4 UTSW 1 82507377 splice site probably null
R1217:Col4a4 UTSW 1 82466730 critical splice donor site probably null
R1465:Col4a4 UTSW 1 82475543 splice site probably null
R1465:Col4a4 UTSW 1 82475543 splice site probably null
R1474:Col4a4 UTSW 1 82458207 nonsense probably null
R1508:Col4a4 UTSW 1 82433557 missense unknown
R1640:Col4a4 UTSW 1 82513491 missense unknown
R1678:Col4a4 UTSW 1 82464380 missense unknown
R1827:Col4a4 UTSW 1 82517709 missense unknown
R1930:Col4a4 UTSW 1 82444321 splice site probably null
R1931:Col4a4 UTSW 1 82444321 splice site probably null
R2092:Col4a4 UTSW 1 82476667 missense unknown
R2122:Col4a4 UTSW 1 82434592 missense unknown
R2132:Col4a4 UTSW 1 82475581 missense unknown
R2396:Col4a4 UTSW 1 82484793 missense unknown
R2418:Col4a4 UTSW 1 82510657 missense unknown
R2679:Col4a4 UTSW 1 82507332 missense unknown
R3085:Col4a4 UTSW 1 82507285 critical splice donor site probably null
R3437:Col4a4 UTSW 1 82474889 missense unknown
R3697:Col4a4 UTSW 1 82518958 missense unknown
R3730:Col4a4 UTSW 1 82433472 splice site probably null
R3752:Col4a4 UTSW 1 82458215 missense probably damaging 0.97
R4085:Col4a4 UTSW 1 82448909 critical splice donor site probably null
R4087:Col4a4 UTSW 1 82501643 missense unknown
R4088:Col4a4 UTSW 1 82501643 missense unknown
R4090:Col4a4 UTSW 1 82501643 missense unknown
R4213:Col4a4 UTSW 1 82430865 missense unknown
R4422:Col4a4 UTSW 1 82467559 missense unknown
R4596:Col4a4 UTSW 1 82448940 missense unknown
R4755:Col4a4 UTSW 1 82518895 missense unknown
R4757:Col4a4 UTSW 1 82506187 missense unknown
R4793:Col4a4 UTSW 1 82516820 missense unknown
R4812:Col4a4 UTSW 1 82439874 missense unknown
R4833:Col4a4 UTSW 1 82507323 missense unknown
R5259:Col4a4 UTSW 1 82431614 missense unknown
R5264:Col4a4 UTSW 1 82471312 missense unknown
R5265:Col4a4 UTSW 1 82471312 missense unknown
R5281:Col4a4 UTSW 1 82471312 missense unknown
R5283:Col4a4 UTSW 1 82471312 missense unknown
R5284:Col4a4 UTSW 1 82471312 missense unknown
R5387:Col4a4 UTSW 1 82471312 missense unknown
R5388:Col4a4 UTSW 1 82471312 missense unknown
R5435:Col4a4 UTSW 1 82431728 missense unknown
R5534:Col4a4 UTSW 1 82465238 missense unknown
R5666:Col4a4 UTSW 1 82463300 critical splice donor site probably null
R5670:Col4a4 UTSW 1 82463300 critical splice donor site probably null
R5943:Col4a4 UTSW 1 82502737 missense unknown
R5996:Col4a4 UTSW 1 82433449 missense unknown
R5999:Col4a4 UTSW 1 82470340 missense unknown
R6112:Col4a4 UTSW 1 82431604 missense unknown
R6192:Col4a4 UTSW 1 82462151 missense probably damaging 1.00
R6237:Col4a4 UTSW 1 82484752 missense unknown
R6419:Col4a4 UTSW 1 82444207 critical splice donor site probably null
R6458:Col4a4 UTSW 1 82433546 missense unknown
R6460:Col4a4 UTSW 1 82444253 missense unknown
R6481:Col4a4 UTSW 1 82431499 missense unknown
R6522:Col4a4 UTSW 1 82465304 missense unknown
R7000:Col4a4 UTSW 1 82475051 missense unknown
R7015:Col4a4 UTSW 1 82484671 missense unknown
R7055:Col4a4 UTSW 1 82496757 missense unknown
R7288:Col4a4 UTSW 1 82470184 missense unknown
R7293:Col4a4 UTSW 1 82501664 missense unknown
R7300:Col4a4 UTSW 1 82464361 missense unknown
R7458:Col4a4 UTSW 1 82476669 missense unknown
R7520:Col4a4 UTSW 1 82484808 nonsense probably null
R7727:Col4a4 UTSW 1 82506514 missense unknown
R7803:Col4a4 UTSW 1 82467419 critical splice donor site probably null
R7953:Col4a4 UTSW 1 82431689 missense unknown
R7959:Col4a4 UTSW 1 82484780 missense unknown
R7982:Col4a4 UTSW 1 82549162 start gained probably benign
R8000:Col4a4 UTSW 1 82519018 missense unknown
R8057:Col4a4 UTSW 1 82501591 missense unknown
R8126:Col4a4 UTSW 1 82431007 missense unknown
R8406:Col4a4 UTSW 1 82501611 missense unknown
R8699:Col4a4 UTSW 1 82433455 missense unknown
R8835:Col4a4 UTSW 1 82447313 missense unknown
R8916:Col4a4 UTSW 1 82501667 missense unknown
R8921:Col4a4 UTSW 1 82431533 missense unknown
R8990:Col4a4 UTSW 1 82473555 missense unknown
R9002:Col4a4 UTSW 1 82449032 missense probably benign 0.26
R9116:Col4a4 UTSW 1 82431752 missense unknown
R9176:Col4a4 UTSW 1 82463349 missense unknown
R9211:Col4a4 UTSW 1 82506501 missense unknown
R9246:Col4a4 UTSW 1 82430956 missense unknown
R9463:Col4a4 UTSW 1 82431076 missense unknown
R9666:Col4a4 UTSW 1 82496670 missense unknown
R9686:Col4a4 UTSW 1 82474962 missense unknown
R9705:Col4a4 UTSW 1 82465313 missense unknown
R9749:Col4a4 UTSW 1 82463353 missense unknown
R9774:Col4a4 UTSW 1 82484665 critical splice donor site probably null
X0020:Col4a4 UTSW 1 82517673 critical splice donor site probably null
Z1088:Col4a4 UTSW 1 82430917 missense unknown
Mode of Inheritance Autosomal Recessive
Local Stock Sperm, gDNA
MMRRC Submission 034328-MU
Last Updated 2016-05-13 3:09 PM by Stephen Lyon
Record Created unknown
Record Posted 2010-02-04
Other Mutations in This Stock Stock #: N/A - 293 Run Code:
Validation Efficiency: 165/193

GeneSubstitutionChr/LocMutationPredicted EffectZygosity
Exoc1l T to C 5: 76,664,339 (GRCm39) S143P probably benign Homo
Gm7634 T to A 1: 16,124,084 (GRCm39) noncoding transcript Homo
Kras T to A 6: 145,177,940 (GRCm39) M111L probably benign Homo
Lrig1 T to C 6: 94,586,068 (GRCm39) T707A probably benign Homo
Mycbp2 A to G 14: 103,461,898 (GRCm39) probably benign Homo
Or2a7 C to T 6: 43,151,493 (GRCm39) T191I probably benign Homo
Smarcad1 T to A 6: 65,051,898 (GRCm39) F344I probably benign Homo
Sprr4 G to A 3: 92,407,650 (GRCm39) Q51* probably null Homo
Zeb1 A to T 18: 5,767,076 (GRCm39) H529L possibly damaging Homo
Phenotypic Description
Figure 1. Elevated blood urea nitrogen levels in aboa mice. Levels of blood urea nitrogen in wild type and aoba mice at the indicated ages.
Figure 2. Focal segmental glomerulosclerosis in aoba mice. Serial kidney sections were stained with PAS (A-C) or Trichrome (D-F) and imaged at 100x (A and D) or 200x (B-C, E-F). (A) Pitting of the cortical surface, crowding of glomeruli, tubular atrophy, and dilated tubules in 5 month-old aoba mouse. (B) Completely sclerosed glomeruli (arrows) in 5 month-old aoba mouse. (C) Normal staining in 5 month-old aoba mouse. (E) Scarring extending into medulla in 5 month-old aoba mouse. (F) Normal staining in 5 month-old wild type mouse.

The aoba phenotype was identified while screening G3 mice homozygous for random mutations induced by N-ethyl-N-nitrosourea (ENU) for susceptibility to non-lethal doses of mouse cytomegalovirus (MCMV) (MCMV Susceptibility and Resistance Screen).  Spontaneous death was observed in one pedigree.  Homozygous aoba mice are generally healthy and fertile, although slightly leaner than wild type animals.  However, they spontaneously die between six to seven months of age with death occurring under specific pathogen free (SPF) conditions.  Moribund animals display dramatically elevated levels of blood urea nitrogen (BUN) and creatinine (Figure 1), and small, pale kidneys with focal glomerusclerosis (Figure 2), consistent with end-stage renal disease (ESRD) and kidney failure.  Proteinuria (Table I), hematuria (Table II), and leukocytes in the urine (Table III) could be detected in aoba mice as early as two months of age.  In addition, these animals exhibited enlarged stomachs containing undigested food (1)

Table I. Proteinuria in wildtype and Aoba mice

Proteina

2 months

3 months

4 months

5 monthsb

Negative

1

14

0

0

Trace

3

6

0

1

(+) or 30 mg/dL

1

2

5

1

(++) or 100 mg/dL

0

1

2

2

(+++) or 500 mg/dL

0

0

0

0

Total number of mice analyzed

5

23

7

4

aAll wildtype mice (2-5 per age group, 14 total) were negative for proteinuria.

bUrine could not be collected from 2 additional 5-month-old mice.

Table II. Hematuria in wildtype and Aoba mice

Blooda

2 months

3 months

4 months

5 months

Negative

1

10

2

0

Trace

4

8

4

2

About 50 Ery/uL

0

4

0

0

About 250 Ery/uL

0

0

0

2

Total number of mice analyzed

5

22

6

4

aAll wildtype mice (2-5 per age group, 13 total) were negative for hematuria.

Table III. Presence of leukocytes in urine from wildtype and Aoba mice

Leukocytesa

2 months

3 months

4 months

5 months

Negative

0

8

0

0

Trace

2

8

0

1

(+)

3

4

5

3

(++)

0

2

1

0

Total number of mice analyzed

5

22

6

4

aAll wildtype mice (2-5 per age group, 13 total) were negative for the presence of leukocytes in the blood.

Figure 3. Sensorineural hearing loss in aoba mice. (A) Representative ABR recordings in response to click stimuli in a 5 month-old wild type mouse and two 5 month-old aoba mice, one of which (mutant 1) had a normal ABR threshold. (B) Scatter plot showing the ABR thresholds for individual 5 month-old wild type and aoba mice on a pure C57BL/6J background. Each point represents data from one mouse, and the bar indicates the mean of all values.

Some aoba mice display moderate sensorineural hearing loss with an increased threshold for auditory brain stem response (ABR), but this phenotype has low penetrance (Figure 3).

Nature of Mutation
Figure 4. Identification of a Col4a4 mutation in aoba mice. (A) Trace file showing a G to A transition destroying the specific donor site in intron 8 of the Col4a4 gene. (B) Proteinuria was mapped to chromosome 1 using bulk segregation analysis. Homozygous aoba males were outcrossed to C57BL/10L females, the F1 progeny were intercrossed, and proteinuria was measured in 4 to 5 month-old F2 mice. BSA was performed with genomic DNA from 17 mutant and 28 control F2 mice. The combined LOD score versus 124 autosomal SNP markers that distinguish C57BL/6J and C57BL/10J mice is shown.

Whole genome sequencing of a homozygous aoba mouse using the SOLiD technique was performed (1).  Briefly, genomic DNA was sheared into 100 to 110 bp fragments, ligated to P1 and P2 adaptors, and amplified using library primers. The resulting short fragment DNA library was then clonally amplified onto SOLiD P1 DNA beads and templated beads were loaded onto two slides and subjected to SOLiD sequencing through an extension length of 50 bp fragment reads. Reads were mapped to the reference genome (C57BL/6J assembly Build 37) with a maximum of 6 mismatches allowed for each read. After subtracting mismatches, 88.5%, 82.5%, and 74.56% of coding/splicing sequence was covered at least 1X, 2X or 3X, respectively.  Validation sequencing using the Sanger method was attempted on all nucleotides for which discrepancies were seen at 3x or greater coverage, with 132 of 146 discrepancies successfully processed.

Whole genome SOLiD sequencing followed by Sanger sequencing validation identified a total of 11 mutations in aoba mice, one of which occurred in the Col4a4 candidate gene (Figure 4A).  This was a G to A transition at position 82,532,315in the Genbank genomic region NC_000067 for the Col4a4 gene on chromosome 1 (GTAAGTTTTA -> ATAAGTTTTA).  The mutation is located within intron 8, one nucleotide downstream from the previous exon 8, and impairs the donor splice site of intron 8.  Reverse transcription and PCR amplification revealed a deletion of 8 nucleotides within the aoba Col4a4 cDNA, resulting in a frame-shift, the addition of 13 aberrant amino acids after residue 176, and a premature stop codon.  Col4a4 contains 48 exons.  Multiple Col4a4 transcripts are displayed on Ensembl.

      <--exon 7         <--exon 8 intron 8--> exon 9--> <--exon 10

47754 GGCCCCCCT……GTTAAAGGT ATTCAG GTAAGTTTT………GGAGACCG…………GGATCTAG 55167

153   -G--P--P-……-V--K--G- -I--Q-               -R--P-………………-I--*- 189

             correct       deleted                   aberrant

The donor splice site of intron 8, which is destroyed by the aoba mutation, is indicated in blue lettering; the mutated nucleotide is indicated in red lettering.  The 8 deleted nucleotides are highlighted in gray.

In order to validate that the aoba mutation was causing end stage renal disease in aoba mice, mapping was performed by outcrossing homozygous animals to the closely related C57BL/10J strain and intercrossing F1 animals to yield F2 mice.  Proteinuria was measured in four to five month-old animals and the DNA samples from 17 phenotypically mutant and 28 phenotypically normal animals were isolated and quantitated by real-time PCR.  Equal amounts of DNA from each mutant and WT mouse were added to separate pools and subjected to bulk segregation analysis (BSA) using a total of 127 SNPs that distinguish the C57BL/6J and C57BL/10J strains.  C57BL/6J and C57BL/10J allele frequencies were calculated based on normalized capillary sequencing chromatogram peak heights, which reflect the quantity of a given nucleotide at each position in the DNA sequence (2).  BSA localized the aoba mutation to a ~43 Mb region on chromosome 1, with LOD scores of 7.1, 8.9, and 10 at markers 75,483,331; 95,571,814; and 118,565,405, respectively (Figure 4B).  This result was confirmed by genotyping individual F2 mice for the C57BL/6J and C57BL/10J alleles of each of these markers. The LOD scores calculated from this analysis were 5.9, 10.2, and 10.2 at markers 75,483,331; 95,571,814; and 118,565,405, respectively.  As the Col4a4 mutation is located at 82,532,315 on chromosome 1, it is likely to be responsible for kidney disease in aoba mice.  The adjacent Col4a3 candidate gene was excluded by sequencing (1).
Illustration of Mutations in
Gene & Protein
Protein Prediction
Figure 5. Domain structure of the COL4A4 protein. COL4A4 is a member of the type IV collagen family and contains three structurally distinct domains: an N-terminal domain rich in cysteine and lysine residues (known as the 7S domain), a collagenous domain largely composed of Gly-Xaa-Yaa repeats, and a C-terminal noncollagenous (NC1) domain.
Figure 6. Formation of the α3 α4 α5 collagen network. Collagen monomers first associate through their NC1 domains, followed by formation of the triple-helical structure of the collagenous domains, and covalent bonding between the 7S domains. The collagen trimers then form a collagen network by dimerizing through their NC1 domains and forming tetramers through the 7S domains.

The Col4a4 gene encodes a 1682 amino acid member of the type IV collagen family. In vertebrates, the collagen superfamily contains 28 different types of collagen (3;4). The type IV collagen family includes six highly homologous α-chains, α1-α6. Each chain contains three structurally distinct domains; an N-terminal domain of about 140 amino acids rich in cysteine and lysine residues (known as the 7S domain), a collagenous domain of about 1300 residues largely composed of Gly-Xaa-Yaa repeats, and a C-terminal noncollagenous (NC1) domain that is roughly 230 amino acids long (Figure 5). Based on sequence homology, these six α-chains can be divided into two groups; the α1, α3, and α5 chains, and the α2, α4, α6 chains. Type IV collagen α-chains form heterotrimers to make up single collagen molecules. In turn, collagen heterotrimers dimerize through their NC1 domains and form tetramers through their 7S domains to form a collagen network (3;5-7) (Figure 6).    

Type IV collagen α-chains are processed into mature form and secreted extracellularly where they form collagen networks. Like other collagens, collagen IV undergoes many posttranslational modifications (4). These include removal of the signal peptide in the endoplasmic reticulum (ER), hydroxylation of many proline and lysine residues, glycosylation of specific hydroxylysine residues, addition of mannose-rich oligosachharide and the formation of important intrachain disulfide bonds between 12 conserved cysteine residues in the NC1 domain. Biosynthesis of collagen is highly dependent on at least one specific molecular chaperone, Hsp47 (8;9). In collagen IV, N-linked oligosaccharides are located in the 7S domain and hydroxylysine linked disaccharides are scattered along the entire molecule. These posttranslational modifications likely provide specificity to the alignment and assembly of collagen IV networks (10).   
 
In the Gly-Xaa-Yaa repeats found in the large collagenous domain, X and Y are often proline and hydroxyproline residues. These sequences have a high propensity to form supercoiled triple helical structures (11). Type IV collagen collagenous domains contain numerous interruptions in the repeat motifs that provide molecular flexibility to the collagen networks, and may provide sites for cell-binding and interchain crosslinking (12). All type IV collagen α-chains contain numerous RGD motifs that are known to be integrin-binding sites. However, extensive data on the binding of various integrins to α1, α2, and α3 suggest that these interactions are mostly RGD-independent (3;13;14)  
 
Figure 8. The quarternary organization of the α3 α4 α5 hexamer. The α3 chain is shown in cyan, the α4 chain in blue, and the α5 chain in purple. In the top diagram, the dimerizing NC1 domains are indicated by the colored ovals. The cystal structure of the NC1 hexamer is shown in the bottom picture. Quaternary model is from Borza et al. J Biol Chem. 277, 40075-40083 (2002). UCSF Chimera model is based on PDB 1LI1, Than et al. Proc. Natl. Acad. Sci. USA. 99, 6607-6612 (2002). This image is interactive. Click on the 3D structure to view it rotate.
The six distinct type IV collagen α-chains interact to form three different heterotrimers; α1 α1 α2, α3 α4 α5, and α5 α5 α6 (3;15;16), directed by recognition sequences in the NC1 domains (15;17). The NC1 domains within each trimer “swap” a β-hairpin structure into a four-stranded antiparallel β-sheet docking site located on the neighboring NC1 domain. This domain swapping is the basis for the high degree of specificity involved in the assembly of the heterotrimers. α2-like NC1 domains are the driving force behind heterotrimer assembly as the α2 NC1 domain has a higher affinity for forming a heterodimer with an α1 NC1 domain than with itself (18). It is likely that the α4 NC1 domain plays a similar role during the formation of the α3 α4 α5 heterotrimer. Once assembled in the ER, collagen heterotrimers dimerize in an end-to-end conformation through their NC1 domains, which form a hexamer (19;20). The hexamer is stabilized through extensive hydrophobic and hydrophilic interactions, and are covalently linked by the formation of intermolecular thioether crosslinks formed between specific methionine and hydroxylysine or lysine (in the case of the α4 chain) residues residing within apposing NC1 domains (21;22). Two α3 α4 α5 heterotrimers form dimers by linking the α3 and α5 NC1 domains. The α4 chain only forms a covalent bond with the other α4 chain on the apposing heterotrimer (23) (Figure 7).
 
Of all the α-chains, α4 contains the most interruptions in the collagenous domain with a total of 26, and in addition contains the most cysteine residues in its collagenous and 7S domains. Thus, the α3 α4 α5 collagen network has distinct biological properties from the other type IV collagen networks due to a larger number of inter- and intrachain disulfide bonds (24) (see Background). 
 
The aoba mutation destroys the donor splice site of intron 8, deletes 8 nucleotides from the Col4a4 cDNA, and results in premature protein truncation. 
Expression/Localization
In mice, the collagen IV α4 chain is highly expressed in kidney and lung and detected at lower levels in heart, muscle and skin. Colocalization with α3 and α5 occurs in the glomerular basement membrane (GBM), and tubular basement membrane (TBM) of the kidney, as well as synaptic basal lamina (25). α3, α4, and α5 are expressed in and secreted by the podocytes of the kidney (26). α4 is expressed in the conjunctival basement membrane of the eye (27), in some of the basement membranes of developing tooth buds (28), and is found in the inner ear (29). Examining the collagen IV expression patterns during the development of several tissues in the mouse reveals that the basement membranes of several tissues undergo a developmental shift in collagen expression. During early development of the lens capsule, the testis, and the GBM, the type IV collagen expressed in these tissues is α1 and α2. As these tissues differentiate, the α1 α1 α2 network is replaced with α3, α4, and α5 (25;30-32). In dogs, a similar switch has been noted during inner ear development (33)
 
In humans, Northern blot analysis found high expression of COL4A4 mRNA in adult human kidney, skeletal muscle, and lung, and in fetal kidney and lung with weaker expression in fetal heart. Expression of COL4A4 largely overlapped that of COL4A3 (34). α4 is also expressed in the retinal pigment epithelial basement membrane and the lens capsule (32;35), as well as the inner ear (36). As in mice, there is a developmental switch in collagen expression in certain basement membranes from α1 and α2 to α3, α4, and α5 (30;32;37).
Background

The basement membrane (BM) is a highly specialized extracellular matrix (ECM) containing collagen IV, laminins, nidogen and sulfated proteoglycans that lies beneath epithelial and endothelial cells and surrounds muscles, nerves, adipocytes and smooth muscle cells.  The laminin polymers found in the BM function as the initial template for assembly (38). In the trans-Golgi network (TGN), collagen type IV chains form heterotrimeric protomers.  These protomers are then secreted extracellularly, where they associate to form a supramolecular structure, mediated by cell surface integrins and/or by binding interactions with laminin and other BM-associated macromolecules (3;5).  The highly ordered and crosslinked nature of type IV collagen networks endows BMs with their structural integrity (5).  The BM has a wide variety of biological activities such as providing adherent cells with structural support and functional cues through its biomechanical properties, and the presence of adhesion receptor ligands and matrix-bound growth factors.  BMs also prevent the diffusion of all but the smallest molecules. However, normal and cancerous cells can interact via integrin cell surface receptors to mediate cell adhesion and initiate tissue-invasive mechanisms including the expression of ECM-degradative proteases such as certain matrix metalloproteases (5)

Figure 9. A, The nephron is the basic functional unit of the kidney. The nephron's initial filtering component is the glomerulus, a tuft of capillaries responsible for filtering blood to form urine. B, The glomerular basement membrane contains containing collagen IV and is lined on one side by the fenestrated endothelium of the capillaries and on the other side by a specialized epithelium facing the urinary space. The specialized epithelium contains podocytes with interdigitating foot processes that form filtration slits. These filtration slits are spanned by modified adherens junctions called glomerular epithelial slit diaphragms, which link foot processes and also keep filtration slits open. The glomerular basement membrane and the glomerular epithelial slit diaphragms provide barriers to larger proteins, preventing protein leak into urine.

Basement membranes play an important role in the structure and function of the kidney.  The kidney is the site of plasma filtration, which occurs in the renal glomeruli (Figure 9).  The glomeruli are made up of tufts of capillaries composed of a defined basement membrane known as the glomerular basement membrane (GMB), which is lined on one side by fenestrated endothelium facing the blood and on the other side by a specialized epithelium facing the urinary space.  The specialized cells of the epithelium have long and extensive foot processes, and are known as podocytes.  The glomerular basement membrane (GBM) and the glomerular epithelial slit diaphragms (GESD) provide barriers to proteins larger than 70 kDa.  The GESD is a modified adherens junction between the foot processes of the podocyte, and may be the most important barrier for preventing protein leak into the urine (39).

Alport syndrome (OMIM #301050; #203780) is an inherited progressive disease of the kidneys, often associated with sensorineural deafness and occasionally ocular lesions.  The renal disease becomes evident as recurrent microscopic or gross hematuria, sometimes in childhood, and progresses to end-stage renal disease (ESRD).  Patients can exhibit extensive glomerusclerosis.  Alport syndrome shows considerable clinical heterogeneity, with X-linked forms of the disease caused by mutations in the COL4A5 gene, and autosomal recessive and dominant forms caused by mutations in COL4A3 and COL4A4 (40-42).  Heterozygous mutations in COL4A3 and COL4A4 also result in autosomal dominant benign familial hematuria (BFH; OMIM #141200) or thin basement membrane nephropathy (TBMN), which presents with microscopic hematuria, but only occasionally progresses to kidney malfunction (41;43).  Both Alport syndrome and BFH can be considered to be genetic diseases of the glomerular basement membrane (GBM) of the kidney, with Alport patients displaying variable thinning and thickening of the GBM, combined with multilamellation of the lamina densa.  Individuals diagnosed with BFH are characterized by less severe changes in the GBM (42).  Since the α3, α4, α5 chains form a trimeric molecule, a mutation in any of these chains usually results in complete absence of the α3 α4 α5 collagen IV network (42;44)

Animal models of Alport syndrome include spontaneous mutations found in the Col4a5 gene in dogs (45;46), as well as mouse knockouts of Col4a3 and Col4a5 (47-49).  An additional mouse containing a transgenic insertion that results in a genomic DNA deletion encompassing portions of the Col4a3 and Col4a4 genes and their shared promoter, has been described (50).  All of these animal models exhibit GBM abnormalities progressing to end-stage renal disease, mirroring the phenotypes displayed in human patients.  Like the human disease, the course of disease progression and age of kidney failure in these animal models can vary greatly.  C57BL/6J Col4a3?/- mice, compared with mice on a 129/SvJ background, have a significantly delayed disease onset with end-stage renal failure occurring at 194 versus 66 days on average (51).  The difference in disease progression can be partially attributed to the ectopic deposition of the collagen IV α5 and α6 chains in the GBM of knockout mice on the C57BL/6J background, whereas only the α1 and α2 chains were present in the GBM of Col4a3 knockouts on the 129/SvJ background (52)

The initial function of the GBM in both humans and animals with Alport syndrome appears to be retained, likely due to the persistence of an embryonic-like α1 α1 α2 collagen network in place of α3 α4 α5 (30).  Differences between the physical and biological properties of the α1 α1 α2 network vs. the α3 α4 α5 network are likely to underlie the inability of the former to compensate for the lack of the latter.  The α3 and α4 chains are more cysteine-rich than α1 and α2, allowing loop structures to form between triple helical collagen molecules.  Thus, the α3 α4 α5 may provide more structural integrity to the GBM that is needed for appropriate plasma filtration.  In addition, the α1 α1 α2 network is more susceptible to degradation by proteases it encounters in the serum.  The uneven thickness of the GBM characteristic of Alport syndrome is likely due to gradual proteolytic degradation of the α1 α1 α2 network found in these patients (3;24;42).  The proteolytic degradation of the GBM appears to be enhanced by the increased expression of matrix metalloprotease-12 (MMP12), found in the podocytes of animals and humans with Alport syndrome (53).  Increased expression of MMP2, MMP9, as well as TGFβ1, have been found to contribute to the progressive fibrosis found in patients with Alport syndrome (54;55).  

Mouse knockouts of certain integrin receptors also exhibit similar kidney phenotypes.  Integrin receptors are heterodimers composed of an α- and a β-subunit. 18α- and 8β-subunits have been identified in mammalian cells (3).  The predominant integrin expressed by podocytes is α3β1 (56), and in vivo deletion of the α3 subunit results in a strong glomerular phenotype (57).  Although integrins α1β1 and α2β1 are also expressed by podocytes, their role in normal glomerular development is likely to be less important, as mice null for the α1 or α2 subunit only have a minor glomerular phenotype (14;58).  The αvβ3 integrin is also expressed in podocytes and has been demonstrated to bind to the collagen IV α3 chain.  However, deletion of this integrin in mice did not induce renal abnormalities, although increased expression of integrin αvβ3 is associated with diabetic nephropathy (59).  In addition to integrin receptors, the discoidin domain receptors (DDRs) have also been found to bind to collagens (60), and deletion of DDR1 in mice results in GBM thickening and proteinuria (61).

Other diseases associated with collagen IV include Goodpasture’s syndrome (OMIM #233450), and the rare neurological disease encephaloclastic porencephaly (OMIM #175780).  Encephaloclastic porencephaly is caused by mutations in the COL4A1 gene (62).   Goodpasture’s syndrome is an autoimmune disease caused by autoantibodies directed against the collagen IV α3 chain, and is characterized by a rapidly progressive glomerulonephritis and lung hemorrhage (63).  Two dominant epitopes on the α3 chain have been identified, but do not normally show reactivity to the antibodies unless the NC1 hexamer is dissociated (23).  It is likely that environmental factors may cause this exposure, leading to the development of autoantibodies, and an immune reaction resulting in disruption of the glomerular filtration barrier [reviewed by (3)].  In insects, collagen IV peptide fragments constitute a danger system and stimulate the immune system, particularly fragments containing the integrin-binding motif, RGD (64).  Collagen fragments appear to have functional roles in other systems as well.  The NC1 domains of certain collagens, including the collagen IV α1, α2, α3 and α6 chains, have been discovered to contain anti-angiogenic and anti-tumorigenic properties.  The proteolytic release of these soluble NC1 fragments stimulates migration, proliferation, apoptosis or survival of different cell types, and can even alter morphogenetic events.  These events are likely to be mediated by interactions with various integrin receptors [reviewed by (3;7)].  None of these studies identified a similar role for the collagen IV α4 NC1 domain (65)

The collagen IV genes are arranged in three pairs in a head-to-head orientation, and are transcribed in opposite directions with COL4A1 paired with COL4A2, COL4A3 with COL4A4, and COL4A5 with COL4A6.  This allows for coordinate regulation of the gene pairs in specific tissues.  The LIM homeodomain transcription factor, LMX1b, is known to regulate the expression of Col4a3/Col4a4 in the GBM, and Lmx1b -/- animals and patients with heterozygous mutations in the LMX1B gene display kidney phenotypes similar to those described for Alport syndrome (OMIM #161200) (66).  However, the regulation of the promoter sequences for these genes is not well understood, and differences in mRNA levels and proteins in certain tissues have been reported.  Some examples include the expression of collagen IV α4 protein during early tooth morphogenesis and in the conjunctival basement membrane during eye development, in the absence of the α3 chain (27;28).  Similarly, the α3 protein was reported in the adult corneal basement membrane of the eye where α4 was undetectable (27).  In addition, α3 has several alternatively spliced variants that are expressed in human kidneys, while no splice variants have been reported for any of the other chains (67;68).  The role of these splice variants in the collagen network is unknown. 

Putative Mechanism

The aoba mouse is the first known instance of a causative mutation identified by whole-genome high through-put sequencing.  Aoba mice demonstrate a highly similar phenotype to other mouse and animal models of Alport syndrome (69), with spontaneous death occurring due to renal failure.  Histological, blood and urine analyses revealed the typical phenotypes of humans and animals with Alport syndrome; focal glomerusclerosis and elevated levels of blood urea nitrogen.  End-stage renal disease in these animals occurs at six to seven months of age, a similar time point as the renal failure observed in Col4a3 knockout mice on a C57BL/6J background (51).  Considering the similarity between homozygous aoba mice and other mouse models of Alport syndrome, it is likely that the aoba mutation located in the donor splice site of intron 8 results in a profound loss of Col4a4 function.    

Along with the mutation in Col4a4, homozygous aoba mice also display homozygous mutations in nine additional genes; Gm7271, Gm7634, Kras, Lrig1, Mycbp2, Olfr13, Smarcad1, Sprr4, and Zeb1 (Table 1), and it is possible that other ENU induced mutations exist but were not detected in the course of SOLiD sequencing and/or the validation sequencing that followed.  A number of these genes have been knocked out in mice, and none of the phenotypes have been identified in aoba animals.  It is likely that the function of the proteins encoded by these genes remain at least partially intact in aoba mice.

Gene

Mutation type

Protein and function

Knockout phenotype

References

Gm7271

missense

uncharacterized

none

Gm7634

missense

ribosomal protein

none

Kras

missense

GTPase

embryonic lethal, heart and other developmental defects

(70;71)

Lrig1

missense

leucine-rich repeat protein/ interacts with receptors of the EGFR family

psoriasiform epidermal hyperplasia

(72)

Mycbp2

may affect splice site

E3 ubiquitin-protein ligase

neonatal lethality, defective diaphragm innervation, abnormal brain morphology

(73)

Olfr13

missense

olfactory receptor

none

(74)

Smarcad1

missense

ATP-dependent DNA helicase

retarded growth, impaired fertility, skeletal dysplasias, peri- and postnatal lethality

(75)

Sprr4

nonsense

cornifin/expressed in keratinocytes

none

(76)

Zeb1

missense

zinc-finger transcriptional repressor

neonatal lethality, T cell deficiency, skeletal abnormalities

(77;78)

Zeb1, which is mutated in aoba, encodes a transcriptional repressor (TCF8) of the COL4A3 gene.  In humans, mutations in ZEB1 result in posterior polymorphous corneal dystrophy-3 (PPCD3; OMIM #609141), a rare disorder involving metaplasia and overgrowth of corneal endothelial cells, as well as basement membrane abnormalities in the cornea. The overgrowth coincides with ectopic expression of COL4A3 by corneal endothelial cells, and a binding site for TCF8 was also identified in the COL4A3 promoter (79).  Rare cases of PPCD have also been described in humans with Alport syndrome (80).  In one patient with autosomal recessive Alport syndrome, corneal endothelial cells were attenuated and focally lost (81), a phenotype essentially the opposite of that found in patients carrying ZEB1 mutations.  It is possible that aoba mice carrying a mutation in the Zeb1 gene, may display ectopic or overexpression of Col4a3, which may somehow affect the phenotypes caused by lack of Col4a4 function.  However, the interdependence of the collagen IV α3 α4 α5 collagen network on each α-chain suggests that overexpression of α3 would not compensate for the lack of α4.  In addition, as noted above, it is likely that TCF8 protein function in these animals is somewhat intact considering that they do not exhibit the same phenotypes as those found in Zeb1 knockout animals.  Interestingly, overexpression of α3 and α4 in the lens of transgenic mice resulted in cataracts caused by the accumulation of collagen chains in the secretory pathway, and the activation of the stress signaling pathway known as the unfolded protein response (UPR; see the record for woodrat) (82).

Primers Primers cannot be located by automatic search.
Genotyping

Aoba genotyping is performed by amplifying the region containing the mutation using PCR, followed by sequencing of the amplified region to detect the single nucleotide transition. 

Primers
aoba(F): 5’- ATGAAAACCTTGGACCTCAGGCTG -3’
aoba(R): 5’- GTCTCAGAGGCAAGGCAATCTACAG -3’
 
PCR program
1) 95°C             2:00
2) 95°C             0:30
3) 56°C             0:30
4) 72°C             1:00
5) repeat steps (2-4) 29X
6) 72°C             7:00
7) 4°C               ∞
 
Primers for sequencing
aoba_seq(F): same as aoba(F)
aoba_seq(R): 5’- GGCAATCTACAGCCAGAGACTAAG -3’
 
The following sequence of 434 nucleotides (NCBI Mouse Genome Build 37.1, Chromosome 1, bases 82,532,111 to 82,532,544) is amplified:
 
   atgaaaa ccttggacct caggctgata acaatcaagg agaccataat tccatcccag
tgaaatggct acagtgcttt ttactggtga ctaatatttc atatagaaga gaaaatactg
aaatggcagc ctactttttc ctaagtgaat gtctcctttt cagggtcagc ctggggaaaa
tggagaaaaa ggaagatctg tgtacattac tggtggcgtt aaaggtattc aggtaagttt
tatggccatt gtgcgtttga caaagccatt ttcccaatag gaaacacacg ggcagggatt
ttctataaag tattattgga tgtactgatt tattttattt tactttgcca ttaacctcat
acaacagatg gctctgttcc ccttgtggag aggcatgggc cttagtctct ggctgtagat
tgccttgcct ctgagac
 
Primer binding sites are underlined; sequencing primer binding sites are highlighted in gray; the mutated G is indicated in red.
References

82. Firtina, Z., Danysh, B. P., Bai, X., Gould, D. B., Kobayashi, T., and Duncan, M. K. (2009) Abnormal Expression of Collagen IV in Lens Activates the Unfolded Protein Response Resulting in Cataract. J. Biol. Chem.

Science Writers Nora G. Smart
Illustrators Diantha La Vine
AuthorsCarrie N. Arnold and Bruce Beutler
Edit History
2011-08-26 5:35 PM (current)
2011-08-26 2:07 PM
2011-08-10 3:23 PM
2011-08-10 2:49 PM
2011-08-10 2:47 PM
2011-08-10 2:41 PM
2011-08-10 2:38 PM
2011-08-10 2:30 PM
2011-08-10 2:27 PM
2011-02-08 3:50 PM
2011-01-27 6:12 PM
2011-01-07 7:39 AM
2011-01-05 4:27 PM
2011-01-05 4:25 PM
2011-01-05 4:24 PM
2011-01-04 5:10 PM
2011-01-04 5:09 PM
2011-01-04 8:48 AM
2011-01-04 8:48 AM
2011-01-04 8:46 AM
2011-01-04 8:43 AM
2010-12-30 3:33 PM
2010-12-30 2:53 PM
2010-12-30 2:51 PM
2010-12-30 2:51 PM
2010-12-30 2:50 PM
2010-12-30 2:49 PM
2010-12-30 2:47 PM
2010-12-30 2:46 PM
2010-12-30 2:45 PM
2010-12-30 2:20 PM
2010-12-08 3:28 PM
2010-11-08 9:44 AM
2010-10-20 8:10 AM
2010-10-05 12:07 PM
2010-10-05 12:05 PM
2010-10-05 12:04 PM
2010-10-05 12:01 PM
2010-07-30 1:29 PM
2010-02-05 10:39 AM