OUP user menu

Zaspopathy in a large classic late-onset distal myopathy family

R. Griggs, A. Vihola, P. Hackman, K. Talvinen, H. Haravuori, G. Faulkner, B. Eymard, I. Richard, D. Selcen, A. Engel, O. Carpen, B. Udd
DOI: http://dx.doi.org/10.1093/brain/awm006 1477-1484 First published online: 2 March 2007


Distal myopathies have been associated with mutations in titin, dysferlin, GNE, desmin and myosin. Of these, only titin mutations were previously known to cause dominant late-onset distal myopathy. Recent findings, however, have indicated that patients affected with myofibrillar myopathy have a more distal than proximal muscle phenotype and a proportion of these may have mutations in myotilin, ZASP or filamin C, besides previously known desmin and αB-crystallin. Here we report that the disorder in one of the well-characterized autosomal dominant distal myopathy families, the Markesbery et al. family, first reported in 1974, is caused by ZASP mutation A165V. Previous linkage to the titin locus 2q31 proved incorrect. ZASP expression by immunoblotting shows normal presence of the main 32 and 78 kDa bands and immunohistochemistry in patients reveals normal Z-disc localization except for moderate accumulations together with myotilin, desmin αB-crystallin and α-actinin. Muscle imaging reveals involvement in both the posterior and anterior compartments of the lower leg and considerable affection of proximal leg muscles at later stages. Haplotype studies in this family and in five other unrelated families with European ancestry carrying the identical A165V mutation share common markers at the locus suggesting the existence of a founder mutation.

  • myofibrillar myopathy
  • distal myopathy
  • ZASP


Distal myopathy has been recognized since the time of Gowers (Gowers, 1902) or earlier (Linsmeyer, 1895) and refers to disorders which characteristically present with weakness in hands and/or feet as opposed to the more common proximal myopathies. Early descriptions had insufficient data to exclude disorders such as myotonic dystrophy or Charcot–Marie–Tooth disease. Welander first defined a distal disorder that was clearly a myopathy (Welander, 1951). Other distal myopathies were subsequently documented with careful neuropathology including: autosomal dominant late-onset distal myopathy (Markesbery et al., 1974); autosomal recessive distal myopathy (Miyoshi et al., 1977); autosomal recessive distal myopathy with rimmed vacuoles (Nonaka et al., 1981); tibial muscular dystrophy (Udd et al., 1993) and others (Udd and Griggs, 2004). As myopathies have been defined in terms of their molecular genetics (Table 1), it has become clear that myopathies defined by histopathological characteristics can have pronounced distal presentations, particularly the myofibrillar myopathies (MFM) (Selcen et al., 2004). The recent discovery of the molecular defects in many myofibrillar myopathies has prompted reassessment of earlier reported kindreds.

View this table:
Table 1

Distal myopathies with genetic definition

TypeOMIM no.OnsetGenetics
Refer.Age (years)Initial weaknessFamGene & locus
Early onset distal myopathy Laing, MPD11605001–25Anterior lower legADMyosin MYH7,14q
Tibial muscular dystrophy Udd, TMD600334>35Anterior lower legADTitin TTN, 2q31
Distal desminopathy (myofibrillar myopathy)60141915–40Distal leg and forearm + cardiomyopathyADDesmin 2q35
Distal myotilinopathy (myofibrillar myopathy)60920040–60Lower legs and handsADMyotilin 5q31
Zaspopathy (myofibrillar myopathy)60945240–60Lower legs and handsADZASP 10q
Distal MFM (myofibrillar myopathy)608810AdultDistal leg and hands + cardiomyopathyADαB-crystallin 11q
Distal dysferlinopathy Miyoshi, MM25413015–30Posterior lower leg, calfARDysferlin 2p13
Distal myopathy with rimmed vacuoles Nonaka, DMRV60582015–30Anterior lower legARGNE 9p1-q1
Welander distal myopathy, WDM604454>40Hands, finger extensorsAD2p13
Distal myopathy with vocal cord and pharyngeal weakness Feit, MPD260607035–60Asymmetric lower leg and hands + dysphoniaAD5q31
Distal myopathy with pes cavus and areflexia Servidei60184615–50Anterior + posterior lower leg, dysphonia + dysphagiaAD19p13
Adult onset distal myopathy Mahjneh, MPD3Mahjneh et al. (2003)>30Hands or anterior lower legAD8p-q and 12q linked

Z-band alternatively spliced PDZ-motif containing protein (ZASP) is a sarcomeric protein expressed in human cardiac and skeletal muscle at the Z-disk. The functions of ZASP are not fully understood but interactions with α-actinin-2 via its N-terminal PDZ domain suggests at least mechanical anchoring of Z-disk attached proteins, such as titin (Au et al., 2004). The PDZ domain recognizes the C-terminal sequence of a calmodulin-like domain in α-actinin by micromolar affinity. The binding site is different from that recognized by titin Z-repeats and a ternary complex arrangement between the three molecules has been proposed (Au et al., 2004). Recently, the ZASP-like motif (ZM) of ZASP was shown to be required and sufficient for interaction with α-actinin, the PDZ domain binding being secondary (Klaavuniemi and Ylanne, 2006). The ZASP gene, ∼66 300 bp in length, consists of 18 exons, which are differentially spliced to form several isoforms: exon 4 is expressed, without exon 6 in three cardiac isoforms, and exon 6 is expressed, without exon 4 in skeletal muscle isoforms (Huang et al., 2003). Interestingly, both exon 4 and 6 contain the ZM motif, and mutations in exon 4 and 6 have been identified in human-dilated cardiomyopathy (DCM) (Vatta et al., 2003), whereas mutations in exon 6 cause skeletal myopathy (Selcen and Engel, 2005). DCM has also been associated with Z-disk abnormalities involving ZASP mutations in other exons (Knoll et al., 2002; Vatta et al., 2003; Arimura et al., 2004).

The Z-disk forms a multiprotein complex of the sarcomere with mechanical functions to keep the sarcomere structure and the myofilaments in register. Besides the transmission of force during contraction, the Z-disk also has cell signalling functions (Frank et al., 2006). Among the many partners of the major Z-disk component α-actinin, ZASP is a relatively small protein (Faulkner et al., 1999). The knock-out mouse (ZASP orthologue cypher) shows severe congenital myopathy and cardiomyopathy (Zhou et al., 2001).

Human ZASP skeletal muscle disease causing mutations have so far been described only in 11 patients with MFM (Selcen and Engel, 2005). The term MFM is used for a distinctive histopathology by muscle biopsy: sarcomere disintegration, accumulation of myofibrillar degradation products and ectopic expression of multiple Z-disk proteins and dystrophin. In MFM patients, so far five different molecular genetic causes have been identified: mutations in desmin, myotilin, ZASP, αB-crystallin and filamin C. These five genes combined account for half of identified MFM cases (Selcen and Engel, 2005). The clinical phenotype in patients with MFM is heterogeneous, with variable age of onset, proximal or distal presentation and with variable occurrence of concomitant cardiomyopathy and neuropathy.

The Markesbery et al. (Markesbery et al., 1974, 1977) family has been well-characterized clinically and pathologically both on light microscopy and ultrastructure. After linkage to the titin locus 2q31 was determined in a clinically similar autosomal dominant distal myopathy, tibial muscular dystrophy, linkage studies with these markers produced significant LOD scores for the 2q31 locus in this family (Haravuori et al., 1998). Subsequent extensive sequencing of the titin gene did not, however, identify a mutation. Because the pathology in the Markesbery et al. family was compatible with MFM, both myotilin and ZASP were sequenced and a previously identified mutation in ZASP was identified in affected family members. This is the first large distal myopathy family reported with zaspopathy. Disease manifestations include late mild–moderate cardiomyopathy in some patients but few or no neurogenic features and no EMG myotonia. We characterize the phenotypic spectrum and we also report on ZASP expression in mutant muscle and identify a distinctive pattern of muscle involvement by magnetic resonance imaging.

Material and methods

Twenty previously unreported individuals have been examined serially over a span of 10–35 years. Following identification of the gene lesion patients were re-examined. As noted in Fig. 1 and Table 2, 10 of the 20 subjects analysed by molecular genetics were affected. Disease symptoms began at age 44–49 years and signs of distal weakness were detected by age 31–49 years. All subjects remained ambulatory (ages 40–70 years). Blood samples and muscle biopsies were used after informed consent of the patients, and all procedures were approved by the ethics committees of the respective institutions according to the Declaration of Helsinki.

View this table:
Table 2

Clinical findings of affected individuals

Subject/case/sexSymptom onset (age in years)Signs onset (age in years)Current status: muscle weaknessCurrent status: muscle atrophy
III-4 (70-year-old woman)Tripping (45)↓ Ankle jerks (39)Distal legs 0–2 Hands 4 Hip flexion and extension 3 Walks with DAFOLower legs, hands and feet interosseus, thenar and hypothenar, minor in distal forearm
III-5 (65-year-old woman)Tripping (44)Distal legs 0–2 Hands 4 Hip flexion and extension 4Lower legs, hands and feet interosseus, thenar and hypothenar, minor in distal forearm
IV-5 (41-year-old woman)None↓ Toe spreading (41)Heel cord contracturesNo
IV-6 (57-year old woman)NoneFoot dorsiflexion 4/4 (45)Foot dorsiflexion 2/2No
IV-8 (55-year old man)Not examined
IV-9 (54-year-old woman)Tripping (49)Foot dorsiflexion (53)Foot dorsiflexion 5-/5-No
IV-10 (52-year-old woman)NoneFoot/toe dorsiflexion (41)Foot dorsiflexion 4/4No
IV-11 (49-year-old woman)NoneNoneNormalNo
IV-12 (45-year-old woman)NoneNormalNormalNo
IV-18 (39-year-old man)NoneNoneNormalNo
Fig. 1

Pedigree of family with late-onset distal myopathy. Individuals analysed in the current study are marked.

Muscle magnetic resonance imaging

Two patients (III-4 and IV-9) were studied with a 1.5 tesla GE Sigma machine. Ten-millimetre slice thickness cross sections from the pelvic girdle, thighs and lower leg muscles were performed with T1-weighted and fat suppression STIR sequences.

Molecular genetic studies

TTN sequencing

TTN was sequenced using DNA of the two affected family members and two healthy relatives of the family (Markesbery et al., 1974). Specific primer pairs were designed, using Primer3 software, to amplify genomic fragments of TTN (Haravuori et al., 2001; Hackman et al., 2002). Sequencing was performed using a Big-Dye Terminator Cycle sequencing Ready Reaction DNA kit (Applied Biosystems Foster City, CA), according to the manufacturer's instructions (Chadwick et al., 1996), and reactions were run on an ABI 377 sequencer (Perkin–Elmer, Applied Biosystems, Foster City, CA). Sequence data were analysed with Sequencing Analysis 3.3 (Perkin–Elmer, Applied Biosystems, Foster City, CA) and Sequencher 3.0 software (Gene Codes Corporation, Ann Arbor, MI). The TTN sequences of the complete M-line TTN (exons Mex1–6), parts of the Z-disc region (exon 7Z rep), N2A line (exons 101–107), I-band (exons I82–83), PEVK region (exons PEVK1-6) and A-band (exons A159–A168) were sequenced in a non-productive search of a disease associated mutation.

Identification of ZASP mutations

All primers were designed using genomic sequence (GenBank accession number AC067750) and ZASP amino acids were numbered according to the protein sequence (GenBank accession number AAH10929). Exons 6 and 9 of the ZASP gene were PCR amplified using genomic DNA from one affected family member of the Markesbery et al. family as template. The sequences of intronic PCR primers were: 5′-TGGGAGATCTCTCTCGACAC-3′ (exon 6 forward), 5′-GTGAGGGAAGAAAGCTGGTC-3′ (exon 6 reverse), 5′-CTCTGCCCCACCTGTTAGAC-3′ (exon 9 forward) and 5′-AGGTTTGGTGGGTACAGAGC-3′ (exon 9 reverse). Amplified fragments were purified and sequenced in both directions using dye terminator cycle sequencing strategy and ABI PRISM 3100 Genetic Analyser (Perkin–Elmer, Applied Biosystems, Foster City, CA).

After identification of the A165V mutation in exon 6 of one family member, solid-phase minisequencing was used to extend the study to the rest of the family (Suomalainen and Syvänen, 1996). Standard PCR amplification of exon 6 was carried out using 0.1 pmol/μl of biotinylated sense primer (5′-CTACCAGGAACGCTTCAACC-3′) and 0.5 pmol/μl of the unbiotinylated antisense primer (5′-GTGAGGGAAGAAAGCT GGTC-3′). Two 10-μl aliquots of each PCR product were transferred on streptavidin-coated microtitration wells (Combiplate 8, Labsystems, Helsinki, Finland), denatured and washed to remove the unbiotinylated product. Two 50-μl minisequencing reactions were performed for each DNA fragment with 10 pmol of the detection primer (5′-GCGATGGCATCCATGATG-3′) and 0.5 U Dynazyme DNA polymerase (Finnzymes) and 3HdATP (Amersham) to detect the mutant and 3HdGTP to detect the wild-type allele. The incorporated label was detected as Optiphase Hisafe (Wallac, Turku, Finland) scintillation (cpm) with a beta counter (Wallac). The ratio between the mutant and normal nucleotide cpm is <0.1 in a homozygote for the wild-type allele, and between 0.5 and 2.0 in a heterozygote for the mutant allele. Some of the minisequencing results were confirmed by PCR amplified genomic DNA sequencing.

A165V mutation haplotype founder study

Haplotype studies were performed on six DNA samples of affected members from six unrelated families of European descent sharing the same A165V mutation in the ZASP gene. The samples analysed were from one affected member of the Markesbery et al. family, one sample of a distal myopathy patient from UK, one of a distal myopathy patient from France, all referred to us as titinopathy candidates and shown to have the A165V mutation. In addition, samples of three patients Pt8, Pt9 and Pt10 (US1, US2 and US3 in Table 3), with previously reported ZASP A165V mutations, were included in the analysis (Selcen and Engel, 2005).

View this table:
Table 3

Haplotyping results of the ZASP locus

Embedded Image

Microsatellite markers

Six polymorphic microsatellite markers on chromosome 10q22.3–q23.22 spanning a region of 5 Mb were used for haplotyping: D10S1739, D10S523, D10S2224, D10S579, D10S215 and D10S1765 (Bowles et al., 2000; Sigma-Genosys, Cambridge, UK). PCR reactions, electrophoresis and genotyping were performed with minor modifications to procedures described elsewhere (Aaltonen et al., 1993). Haplotype analysis was performed using ABI 377 and Genotyper 2.0 software (Applied Biosystems, Foster City, CA).

SNP markers

Using the software tagger implemented in Haploview (http://www.hapmap.org/), Caucasian SNP data from the HapMap project were used to select 8 tag SNPs from a total of 44 SNPs across a short genomic area of 50 kb around the reported mutation A165V in exon 6 of the ZASP gene (LDB3) on chromosome 10q22.3–q23.22. The sequence surrounding each SNP was downloaded directly from dbSNP (http://www.ncbi.nlm.nih.gov/projects/SNP/) and PCR primers were designed using the Primer3 software. Ten to twenty nanograms genomic DNA per SNP were used as template in a standard PCR reaction prior to sequencing using BigDye v3.1 chemistry with detection on ABI 3700-automated DNA sequencer, according to manufacturer's instructions (Applied Biosystems, Foster City, CA, USA). The Phred, Phrap and Consed suite of programs from University of Washington (http://www.phrap.org/) were used to align sequences and to aid detection of variants. The SNP haplotyping was done as an academic collaboration with Dr Charles Mein, Genome Centre Manager, Barts and the London, Queen Mary's School of Medicine and Dentistry, The John Vane Science Centre, London, UK. The primers used will be provided on request.

Immunohistochemistry and Western blotting

Quadriceps biopsies from two patients in the Markesbery et al. family with proven ZASP A165V mutation (III-5 and IV-9) were used for immunohistochemistry (IHC) and Western blotting (WB). Biopsies snap frozen in liquid nitrogen cooled isopentane were used to prepare 6 µm serial sections on SuperFrost+ slides (Menzel GmbH & Co KG, Braunschweig, Germany). IHC stainings were done with Ventana BenchMark automated immunostainer, using iVIEW DAB kit (Ventana Medical Systems, Tucson, AZ), and the following antibodies: mouse polyclonal antibody against ZASP (Faulkner et al., 1999) at 1 : 2000; mouse monoclonal antibody (mAb) αB-crystallin, clone 1B6.1-3G4 (Stressgen, San Diego, CA) at 1 : 4000; mAb desmin, clone DE-R-11 (Novocastra Laboratories, Newcastle-upon-Tyne, UK) at 1 : 500; rabbit polyclonal antibody (pAb) myotilin (Mologni et al., 2005) at 1 : 400; mAb α-actinin, clone EA-53 (Sigma–Aldrich, Saint Louis, MO) at 1 : 30 000; pAb telethonin 1–167 (gift from Prof. M. Gautel) at 1 : 75; pAb dystrophin C-terminus (NeoMarkers, Fremont, CA) at 1 : 400. Fixation with 4% paraformaldehyde (PFA) for 15 min was done prior to αB-crystallin staining; all other slides were stained unfixed. Slides were counterstained with Gill's haematoxylin and eosin (Ventana Medical Systems, Tucson, AZ). Samples for WB were prepared as described earlier (Haravuori et al., 2001), using the same biopsies as for IHC, with two additional healthy control samples (ages 45 and 61 years). SDS–PAGE gels (8%) were run using Bio-Rad Mini-Protean equipment (Bio-Rad Laboratories, Hercules, CA). Proteins were transferred from gels to polyvinylidene difluoride (PVDF) membranes, and labelled with ZASP Ab at 1 : 20 000 for 1 h, and detected using horseradish peroxidase (HRP)-conjugated secondary antibody (DAKO P260, Dako, Glostrup, Denmark) at 1 : 1000, followed by enhanced chemiluminescence (ECL) detection with Bio-Rad Immun-Star kit.


Sequencing the ZASP (LDB3) gene identified a previously reported A165V mutation (Selcen and Engel, 2005) in all clinically affected patients in the family. An additional novel variation a non-pathogenic polymorphism S130L was found in non-affected individuals.

The clinical findings in the follow-up of affected patients and in the patients examined remained consistent with the previously reported features of the disease: initial signs in the anterior compartment of the legs after age 40 years and slow progression to involvement of intrinsic muscles in hands and feet and later mild proximal weakness. Considering the molecular genetic results the penetrance of this mutation appears to be virtually 100% at least by the age of 60 years.

Muscle imaging (Fig. 2) showed considerable involvement of posterior calf muscles. The early changes in a 53-year-old female patient (IV-9) with early tripping and stumbling, but with just minimal ankle dorsiflexion weakness on muscle testing, were fatty degeneration of posterior lower leg muscles, gastrocnemius and soleus (Fig. 2A). Later in the course of the disease (patient III-4, 70 years of age) proximal muscles were affected with mild/moderate fatty degeneration and atrophy of gluteus maximus (Fig. 2D), hamstring, vastus medialis and lateralis muscles (Fig. 2C), besides the severe end-stage replacement in gastrocnemius lateralis, soleus, lateral peroneal and anterior compartment muscles (Fig. 2B). Deep long toe flexors and tibialis posterior were relatively preserved.

Fig. 2

MR muscle imaging. Early findings in a 54-year-old woman (case IV-9) at onset of clinical symptoms (tripping and stumbling) consisted of fatty degenerative changes in the posterior compartment of the legs: gastrocnemius and soleus muscles. These changes did not cause clinical muscle weakness in ankle plantar flexion (2A). In a 70-year-old woman (case III-4) with disease duration of 20 years there is fatty replacement and atrophy of gluteus maximus (2D), moderate to severe changes in semimembranosus, vastus medialis and lateralis and adductor magnus muscles (2C), there is severe end-stage replacement of muscle in gastrocnemius lateralis, soleus, lateral peroneal and anterior compartment muscles (2B). Deep long toe flexors and tibialis posterior muscles were more preserved.

Immunohistochemical studies revealed strong accumulation of myotilin, αB-crystallin, and desmin in the affected muscle fibres (Fig. 3A–C). Surprisingly, only moderate ZASP accumulation was observed using the mouse polyclonal antibody (Fig. 3D), and sarcomeric Z-line labelling was seen in longitudinal sections of intact fibres (not shown). α-Actinin showed only mild aggregation, and another Z-disc component, telethonin, revealed some accumulation in the same abnormal fibres, but the label did not exactly co-localize with accumulated myotilin and desmin (Fig. 3E and F). Dystrophin C-terminus did not consistently localize to the accumulated aggregates, even though occasional punctate aberrant cytoplasmic labelling was observed. Membranous indentations were seen in the affected fibres more frequently (Fig. 3G).

Fig. 3

Immunohistochemical staining of Z-disk proteins performed on serial sections of a banked quadriceps muscle biopsy of ZASP A165V patient (III-5). Specific label (DAB) is brown, haematoxylin counterstain is blue. (A) Myotilin, (B) αB-crystallin, (C) desmin, (D) ZASP, (E) α-actinin, (F) telethonin/T-cap, (G) dystrophin C-terminus. Scale bar 20 µm.

Western blotting of ZASP showed normal amounts of normal sized protein isoforms, with molecular masses of 78 and 32 kDa, in two patients with A165V mutation, and the controls (Fig. 4).

Fig. 4

Western blotting analysis of quadriceps muscle biopsies in two patients and two controls using the mouse polyclonal ZASP antibody. Similar 78 and 32 kDa bands are seen in (1) and (2) healthy controls and (3) ZASP A165V patient (IV-9), (4) ZASP A165V patient (III-5).

As an identical A165V mutation turned up in several unrelated families of European ancestry, a common founder background was investigated. Genotyping was performed using six polymorphic microsatellite markers spanning a region of 5 Mb on chromosome 10q22.3–q23.22, and eight polymorphic SNP markers spanning a region of 50 kb around the mutation. Two additional SNPs were detected during the sequencing, rs2354363 and rs10788522. With the combined data obtained from these analyses we were able to indicate that these samples from different families indeed may share a common short haplotype (Table 3). The SNP results indicated a centromeric border of the haplotype in the intron between exon 1 and 2 of the ZASP gene, in a single patient sample by the homozygote SNP CC (rs19788522) compared to a T in the other samples, and in the Markesbery et al. family sample with a homozygous GG for the SNP rs2354363 instead of a T in two other patient samples. The result was further strengthened by the SNPs rs12569813 and rs2354363s (Table 3). Based on this, the suggestive common haplotype seemed to span over at least 34 kb in the ZASP gene from the intron between exons 1 and 2 over the A165V mutation in exon 6 and further outside the gene. The microsatellite marker D10S579 showed a common allele for all samples and could therefore lie inside the common haplotype, but the telomeric microsatellite marker D10S215 at position 89.45 Mb (∼1 Mb telomeric) varied between the samples and was definitely outside the common haplotype (Table 3).


Distal myopathy in this long-studied family proved to be caused by ZASP A165V mutation. The onset of clinical symptoms is after age 40 years as previously determined by clinical analysis (Markesbery et al., 1974), and while the age of first signs varies and can be in the late 30s, the penetrance of the mutation in this family is 100%, at least by age 60 years. Recent imaging studies identified early changes in posterior lower leg muscles before clinical signs were apparent. The pattern of involvement of the lateral peroneal muscles has not been observed in titinopathy or Welander distal myopathies (Mahjneh et al., 2004). Assessment of the background for the re-occurring A165V mutation by microsatellite and SNP markers in this and five other unrelated families showed an identical short haplotype spanning at least a 34-kb segment of DNA around the mutation indicating this is a founder mutation with a common ancient ancestry.

Currently patients and families with MFM type of myopathology need further molecular genetic studies for the five known genes: desmin, myotilin, ZASP, αB-crystallin and filamin C. Our study suggests that distal myopathies with less consistent myopathology should be assessed for mutations in these genes. What might be the strategy for gene search in these situations? The clinical features of the patients in this family and those of the currently known other zaspopathy patients are not sufficiently distinct to suspect ZASP as the primary candidate gene. Welander distal myopathy usually has onset in long finger extensors but may occasionally have a similar presentation as the patients in this family.

Muscle imaging studies presented here show a pattern of early involvement of posterior calf muscles and late involvement of all lower leg muscles, which is different from the common phenotype in distal myopathy with titin mutations, and from the imaging findings in desminopathy. However, myotilinopathy may cause a similar pattern of muscle involvement (Penisson-Besnier et al., 2006).

Our studies using immunohistochemistry for the MFM proteins show focal sarcoplasmic accumulations in abnormal fibres without loss of normally located ZASP expression in most fibres. As with myotilinopathy, abnormal myotilin aggregation is more prominent than abnormal excessive expression of desmin, αB-crystallin or ZASP. This makes myotilin immunohistochemistry a very sensitive marker for MFM pathology, but distinct immunohistochemistry findings for zaspopathy are not identified. Interestingly, another protein with Z-disk localization, telethonin, showed moderate accumulation in the abnormal population of fibres, but with different intracellular localization, suggesting that in affected muscle telethonin does not belong to the same protein interaction group with the MFM proteins. Semiquantitative measurement of ZASP by Western blotting does not show clear abnormality, in particular no change in isoform expression. Neither are the other Z-disk associated proteins α-actinin, myotilin or αB-crystallin abnormal by Western blot (results not shown). Our findings give no indications on possibilities to separate the different genetic backgrounds further by these protein studies in order to target the candidate gene of interest.

Molecular genetic efforts to find the genetic cause of the disease in this family were conducted for many years, after a misleading marginally significant linkage to the titin locus 2q31. Linkage studies in late-onset disorders, such as zaspopathy are difficult even with full penetrance of the gene by age 60 years. Two generation families are difficult to find, and early clinical diagnosis in 40–60 years old family members risks misclassification. Relying on linkage results is usually, but not always, good enough. When reasonable candidate gene approaches are possible they should be considered. In retrospect, the myopathology findings in this family were the clue to correct the molecular genetic approach and solution of the genetic cause.

Clinical presentation of MFM has been very variable. In this particular family the age of onset and pattern of involvement of muscles proved to be rather uniform within the family, with marginally more severe affection in men compared to women, suggesting either gender difference or exogenic factors such as workload. The very late onset of myofibrillar disintegration starting with the Z-disk, as well as the clear predilection for distal muscles has no definite explanation. The reason for cardiomyopathy not being a regular feature in zaspopathy may be explained by the predominant expression of different isoforms in cardiac and skeletal muscle. Whereas mutations in exons 4, 6, 10 and 15 which are expressed in cardiac muscle isoforms, were associated with dilative cardiomyopathy, the A165V and another mutation A147T in skeletal muscle specific exon 6 cause a myopathy dominated by skeletal muscle involvement. These mutations are adjacent to the ZM motif in exon 6 (Klaavuniemi and Ylanne, 2006). Two mutations reported in DCM are similarly positioned in relation to the ZM motif of cardiac exon 4 (Vatta et al., 2003). The other MFM-gene defects desminopathy and myotilinopathy may cause late-onset disease but also early adult onset forms.

Ultimately myofibres undergoing myofibrillar degeneration are lost and replaced by fat and connective tissue, but the pathomechanisms need comprehensive further studies to understand the molecular interplay within the muscle cell.


The SNP haplotyping was done by Charles Mein, Genome Centre, Barts and the London, Queen Mary's School of Medicine and Dentistry, The John Vane Science Centre, London. Telethonin 1-167 antibody was a generous gift from Prof. M. Gautel, King's College, London. G.F. was supported by Grant GGP04088 from the Telethon Foundation-Italy. O.C. was supported by Sigrid Juselius Foundation and Finnish Heart Association. B.U. was supported by Sigrid Juselius Foundation, the Folkhalsan Research Foundation and the Medical Research Fund of Vasa Central Hospital District.


  • Abbreviations:
    dilated cardiomyopathy
    enhanced chemiluminescence
    horseradish peroxidase
    myofibrillar myopathies
    Z-band alternatively spliced PDZ-motif containing protein
    zasp-like motif


View Abstract