Genotype-Phenotype Correlation in Patients with Familial Mediterranean Fever in East Anatolia (Turkey)

Abstract

Aims: Familial Mediterranean fever (FMF) is an autosomal recessive hereditary disease. Clinical symptoms and findings (phenotypes) seen in this disease are generally fever, abdominal pain, and arthritis. Amyloidosis is also a significant complication. Phenotype-genotype correlations in FMF have not been conclusively resolved. The aims of this study were to find the most frequent mutation/genotype of FMF, as well as to investigate the role of genetic factors on the phenotype and on the development of amyloidosis in a population living in East Anatolia (Turkey). This study included 105 adult patients with FMF. DNA samples were obtained from peripheral blood lymphocytes of the patients. Mutations of the Mediterranean fever (MEFV) gene were analyzed with an FMF Strip Assay test kit (ViennaLab Labordiagnostika GmbH, Vienna, Austria). Patients were separated according to genotypes, and phenotypes were compared statistically by the chi-square test. Results: The most frequent mutation was M694V (53%) and the most frequent genotype was M694V/M694V (26%). In total, 81% of the patients experienced abdominal pain, 76% had fever, and 22% had arthritis. Fever and arthritis were determined in similar ratios to other genotypes (76% and 19%, respectively) in the M694V/M694V genotype (74% and 29%, respectively) (p > 0.50 and p > 0.20, respectively). However, the patients without the M694V/M694V genotype (86%) had a higher abdominal pain ratio than did the patients with the M694V/M694V genotype (67%) (p <0.05). Renal amyloidosis was determined in 33% of both M694V/M694V and M680I(G/C)/M680I(G/C) homozygous groups and in 12% of the heterozygous groups (p < 0.02 and p < 0.00002, respectively). In other words, homozygous groups had higher ratios of renal amyloidosis. Conclusions: The most frequent mutation in FMF was M694V and the most frequent genotype was M694V/M694V. Fever, abdominal pain, arthritis, and renal amyloidosis were determined not only in patients with M694V/M694V genotype but also in other genotypes. Therefore, genotypes may not predict phenotypes in FMF. Renal amyloidosis was seen more frequently in homozygous genotypes.

Introduction

Familial Mediterranean fever (FMF) is an autosomal recessive hereditary disease characterized by recurrent attacks of fever, peritonitis, pleuritis, arthritis, and erysipelas-like erythema. This disease is seen most frequently in Turks, Armenians, Arabs, and Sephardic Jews.

The most serious complication of FMF is renal amyloidosis, which leads to kidney failure and then death. In Turkey, FMF is still a leading cause of renal amyloidosis in high ratio (64%), although this ratio is lower in other peoples (Tuglular et al., 2002). At times, amyloidosis may be the only finding of FMF; this condition has been designated as phenotype II. The International Consortium has declared common mutations related to FMF (i.e., M694V, V726A, M680I, M694I, and E148Q) (Bernot et al., 1998). The frequency of mutations also reportedly changes according to ethnicity. For example, M694V is more commonly seen among Sephardic Jews, Turks, and Armenians, E148Q among European and Turkish patients, M694I among Arabs, and M680I among Armenians (Brik et al., 1999; Touitou, 2001; Yilmaz et al., 2001; Majeed et al., 2002). Reports about any relationship between mutations and symptoms or findings of FMF are contradictory, although some original articles suggested that patients with M694V mutation had more symptoms and findings or more amyloidosis (Majeed, 2002; Yalçinkaya et al., 2000b).

The aims of this study were to determine the most frequent mutation/genotype and to investigate the correlations between phenotypes and genotypes in the patients with FMF in a population living in East Anatolia (Turkey).

Materials and Methods

A total of 105 patients (57 male and 48 female) found to have FMF between January 2002 and January 2004 were included in this study. We evaluated FMF patients who had been followed in Ataturk University, Faculty of Medicine, Erzurum, which is located in the eastern regions of Turkey. In these patients, FMF was diagnosed according to criteria suggested by Livneh et al. (1997). The FMF mutations were investigated in the patients using DNA extracted from peripheral blood lymphocytes according to standard procedures. Mutation identification was performed using a microchip DNA analysis system (Nanogen Inc., San Diego, CA). To determine Mediterranean fever (MEFV) gene mutations in patients, an FMF Strip Assay (ViennaLab Labordiagnostika GmbH, Vienna, Austria) kit was used. The E148Q, P369S, F479L, M680I(G/C), M680I(G/A), 1692del, M694V, M694I, K695R, V726A, A744S, and R761H mutations were investigated using this kit. All of the patients were separated into homozygous, heterozygous, and compound heterozygous groups according to their mutations and genotypes. In the patients separated according to mutations (genotype), the ratios of clinical symptoms and findings (phenotype) were compared statistically by chi-square tests in groups with sufficient numbers of patients. Alternatively, the chi-square test corrected by Yates or Fisher exact chi-square test was used for groups with insufficient numbers. The mean beginning ages of FMF were compared with a Student's t-test in patients with different genotypes. In all tests, p < 0.05 was accepted as significant.

Results

The mean age of onset of FMF was 10.6 ± 8 years. FMF appeared at ≤10 years old in 64%, between 11 and 18 years old in 25%, and >19 years old in 11% of the study population. The mean beginning age was 12.7 ± 8.6 years in patients with the M694V mutation (n = 84) and 9.9 ± 6.7 years in patients without the M694V mutation (n = 21). The difference was not statistically significant (t = 1.398, p < 0.10). In all patients, 19 different genotypes were determined. The most frequent mutation, M694V, was detected in 53% of the carrier chromosomes. Others are shown in Table 1.

Table 1.

Frequencies of Mutations in 105 Patients with Familial Mediterranean Fever

Mutation	No. of alleles	%
M694V	111	53
M680I	25	12
V726A	19	9
E148Q	13	6
Others	21	10
No mutation	21	10
Total	210	100

Genotype distributions in the patients are shown in Table 2. The most frequent genotype was M694V/M694V (26%). Properties of genotypes were homozygous (31%), heterozygous (20%), and compound heterozygous (49%). In these patients, there were complaints of abdominal pain (81%), fever (76%), and arthritis (22%). All of these three findings were most frequently reported for the E148Q/M694V genotype and in 16% of all patients. However, only one finding was determined in patients with the M680I/M680I(G/C) genotype (Table 2).

Table 2.

Phenotype Distribution According to Genotypes in 105 Patients with Familial Mediterranean Fever

		Phenotypes (no. of patients)
Genotypes	Total (n)	Fever	Abdominal pain	Arthritis
M694V/M694V	27	20	18	8
M680I(G/C)/M680I(G/C)	6	5	3	1
M680I/M694V	17	17	17	2
M694V/V726A	15	11	13	2
M694V/E148Q	10	8	9	4
M694V/R761H	2	2	2	0
M694V/K695R	1	0	0	0
M680I/V726A	2	0	2	0
M680I/K695R	1	1	1	0
F479L/V726A	1	1	1	0
A744S/K695R	1	1	0	1
P369S/F479L	1	0	0	0
M694V/?	12	10	11	3
M680I/?	3	1	3	0
E148Q/?	2	0	2	0
V726A/?	1	1	1	1
K695R/?	1	1	0	0
P369S/?	1	0	1	0
F479L/?	1	1	1	1
Total	105	80 (76%)	85 (81%)	23 (22%)

When symptoms of FMF were investigated for the frequently seen M694V mutation, fever was seen in higher ratio in the patients (81%) with the mutation than in the patients (57%) without the mutation (χ² = 5.25, p < 0.05); however, the frequencies of abdominal pain and arthritis were not different in the patients with M694V (83% and 23%, respectively) and in patients without M694V mutation (71% and 17%, respectively) (χ² = 1.544, χ² = 0.125; p > 0.20, p > 0.50, respectively).

Fever and arthritis were found in similar ratios to other genotypes (76% and 19%, respectively) in the M694V/M694V genotype (74% and 29%, respectively) (χ² = 0.149, χ² = 1.267; p > 0.50, p > 0.20, respectively). However, abdominal pain ratio in the patients without the M694V/M694V genotype (86%) was higher than in the patients with the M694V/M694V genotype (67%) (χ² = 4.81, p < 0.05).

Abdominal pain was seen in the highest ratio (100%) in patients with the M680I(G/C)/M694V genotype, whereas it was seen in the lowest ratio (50%) in the M680I/M680I(G/C) genotype. This difference was significant (p < 0.01), meaning that patients with M680I(G/C)/M694V experienced more abdominal pain than other patients.

Renal amyloidosis was confirmed by biopsy in 19% (n = 20) of 105 patients with FMF (Table 3). In these patients with renal amyloidosis, the type II FMF phenotype was determined in 30% (n = 6). Renal amyloidosis was determined in 17% (n = 14) of the patients with the M694V mutation (n = 84) and in 28% (n = 6) of the patients without the M694V mutation (n = 21) (Table 3). No differences were seen in renal amyloidosis ratios in patients with or without the M694V mutation (χ² = 1.544, p > 0.20). Renal amyloidosis was determined in 33% of both M694V/M694V and M680I(G/C)/M680I homozygous groups and in 12% of heterozygous groups (χ² = 6.38, p < 0.02, p < 0.00002, respectively). In other words, homozygous groups had higher ratios of renal amyloidosis.

Table 3.

Distribution Genotypes in the Patients with Renal Amyloidosis or Chronic Renal Failure Due to Familial Mediterranean Fever

		No. of patients
Genotype	Total (n)	Amyloidosis	CRF	ESRD
M694V/M694V	27	9	4	3
M680I(G/C)/M680I(G/C)	6	2	2	2
M680I/M694V	17	2	2	1
M694V/V726A	15	1	1	1
M694V/E148Q	10	0	0	0
M694V/R761H	2	0	0	0
M694V/K695R	1	0	0	0
M680I/V726A	2	0	0	0
M680I/K695R	1	0	0	0
F479L/V726A	1	1	0	0
A744S/K695R	1	0	0	0
P369S/F479L	1	1	1	0
M694V/?	12	2	0	0
M680I/?	3	0	0	0
E148Q/?	2	1	1	1
V726A/?	1	0	0	0
K695R/?	1	0	0	0
P369S/?	1	0	0	0
F479L/?	1	1	1	1
Total	105	20 (19%)	12 (11%)	9 (9%)

CRF, chronic renal failure; ESRD, end-stage chronic renal failure.

In the patients with both chronic renal failure and end-stage chronic renal failure, no differences were seen between homozygous or heterozygous mutation ratios (χ² = 0.481; p > 0.50; Table 3). In renal amyloidosis, the ratios of allele numbers were similar in patients with and without the M694V mutation allele (χ² = 0.488, p > 0.30; Table 4). Therefore, the M694V mutation is apparently related with the formation of amyloidosis in FMF.

Table 4.

Number of M694V Mutation Alleles in Patients with Amyloidosis

	No. of alleles
Mutation	Amyloid (+)	Amyloid (−)
M694V (+)	23	88
M694V (−)	13	65

Discussion

The cloning of the pyrin gene provided the necessary tool to investigate the genetic variation among FMF patients in various ethnic populations. The variability of clinical presentation and severity of the course in some patients has prompted researchers to analyze phenotype-genotype correlations. M694V was the most common mutation found in our study. Touitou (2001) reviewed 1390 Turks, 1301 Sephardic Jews, 706 Arabs, and 278 Armenians with FMF, reported in various studies. Turkish patients had mutations that included M694V (45%), M680I (13%), V726A (11%), M694I (7%), and E148Q (2%). It was reported that the most frequent mutation was M694V in Turks, Sephardic Jews, Arabs, and Armenians (Touitou, 2001). In a recent study, 2180 mutations were determined in 1090 patients with FMF. Mutations were M694V in 51.4%, M680I in 14.4%, and V726A in 8.6% (Tunca et al., 2005). However, this study did not include any patients living in East Anatolia. Ureten et al. (2009) reported that the most frequent mutation was M694V in Turkey. These findings established in Turkish patients were similar to results of our study. In a study that included 41 children with FMF living in East Anatolia, homozygous M694V mutation was found in 29.3%, heterozygous M694V in 22%, V726A in 7.3%, M680I in 4.9%, E148Q in 4.9%, and R761H in 2.4% (Ertekin et al., 2005).

Most people develop symptoms of FMF between the ages of 5 and 15 years, with 90% of those affected becoming symptomatic by age 20. Brik et al. (1999) reported that symptoms of FMF began earlier in patients with the M694V mutation than in other patients. Yalçinkaya et al. (2000a) reported that FMF began at later ages in patients with the M694V homozygous mutation than in other patients. However, we found similar ages on onset for all mutations. Similar to our results, Dusunsel et al. (2008) also reported that no correlation could be found between FMF mutations and age of onset of symptoms.

There are contradictory reports about relationships between fever, abdominal pain, arthritis, and genotype. In our study, fever and abdominal pain were noted in higher frequency in patients with the M680I(G/C)/M694V genotype. However, previous Turkish FMF group studies (Brik et al., 1999; Mimouni et al., 2000; Shinar et al., 2000; Yalçinkaya et al., 2000a; Tunca et al., 2005) did not find any significant difference in fever or abdominal pain incidence among genotypes. However, these studies did not provide any information regarding any relationship between the M680I(G/C)/M694V genotype and fever or abdominal pain. Sakallioglu et al. (2006) and Paut et al. (2000) reported that fever was seen more frequently in patients with the M694V/M694V genotype. Shinawi et al. (2000) compared V726A with M680I mutations for the frequency of abdominal pain and reported a higher ratio in patients with the M680I mutation.

Arthritis is a common and important feature of FMF (Livneh et al., 1997; Tuglular et al., 2002; Uyanik et al., 2009). Brik et al. (1999), Mimouni et al. (2000), Shinar et al. (2000), and Olgun et al. (2005) found a higher ratio of arthritis in patients with the M694V/M694V genotype compared with all other genotypes. However, in agreement with Yalçinkaya et al. (2000a), in this study we found that arthritis ratios were similar among all genotypes.

One of the most significant complications of FMF is amyloidosis, which usually affects the kidneys, resulting in nephrotic syndrome and renal insufficiency that progresses to end-stage renal disease (Livneh et al., 1997; Mimouni et al., 2000; Tuglular et al., 2002; Cetinkaya et al., 2003). We found that the patients with both M694V and M680I homozygous mutations had amyloidosis in higher ratio than the patients with heterozygous mutations. Many investigators have reported an association between the M694V, especially in the homozygous state, and the severity of FMF and development of amyloidosis (Brik et al., 1999; Majeed et al., 2005), whereas other authors failed to find evidence for such an association (Yalçinkaya et al., 2000a). Yalçinkaya et al. (1998) described a group of patients with FMF complicated by Amyloid A amyloidosis who were all heterozygous for the V726A mutation. Pras (1998) reported a single patient with amyloidosis who was homozygous for the V726A mutation. The Turkish FMF study group demonstrated a close relationship between the M694V mutation and disease severity. M694V was the most frequent mutation (51.4%) in their study, and they also demonstrated that simple or compound heterozygous M694V mutations were more frequent than homozygous mutations. In this study, the M694V mutation was found in about 42% of the carrier chromosomes in our patient group, and 28.5% of the patients were heterozygous for the M694V mutation (Tunca et al., 2005). In our series, interestingly, there were two cases of amyloidosis carrying a homozygous M680I mutation. The M680I mutation is common in Armenians, Turks, and Arabs, and amyloidosis is extremely rare in this genotype (Majeed et al., 2005). Although many studies have shown M694V to be the leading mutation for risk of developing amyloidosis, patients with mutations other than M694V are also prone to this complication, and some other factors such as environmental stimuli and modifier genes may have additional effects.

The description of phenotype II is a patient with a positive family history of FMF, who has developed Amyloid A amyloidosis without clinical features of FMF. In phenotype II patients, Dusunsel et al. (2008) reported a correlation between the M694V mutation and both phenotype and amyloidosis. Among our patients, there were six cases with a probable diagnosis of phenotype II.

As a consequence, in our study, although the highest ratio of fever and abdominal pain were found in patients with the M680I(G/C)/M694V mutation and amyloidosis was highest in patients with both M694V and M680I homozygous mutations, the symptoms and findings of FMF or amyloidosis were also seen in patients with other mutations. These results suggest the likelihood that environmental factors and other genetic modifiers may play a role in the occurrence of FMF and its complications (amyloidosis, etc.) in various populations.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Bernot

, da Silva

, Petit

et al. 1998. Non-founder mutations in the MEFV gene establish this gene as the cause of familial Mediterranean fever (FMF) Hum Mol Genet, 7:1317-1325.

Brik

, Shinawi

, Kepten

et al. 1999. Familial Mediterranian fever: clinical and genetic characterization in a mixed pediatric population of Jewish and Arab patients. Pediatrics, 103:70-73.

Cetinkaya

, Odabas

, Selcuk

et al. 2003. Amyloidosis due to familial Mediterranean fever: clinical and laboratory findings in 51 patients. Pain Clinic, 15:333-337.

Dusunsel

, Dursun

, Gündüz

et al. 2008. Genotype-phenotype correlation in children with familial Mediterranean fever in a Turkish population. Pediatr Int, 50:208-212.

Ertekin

, Selimoglu

, Pirim

. 2005. Familial Mediterranean fever in a childhood population in eastern Turkey. Pediatr Int, 47:640-644.

Livneh

, Langevitz

, Zemer

et al. 1997. Criteria for the diagnosis of familial Mediterranean fever. Arthritis Rheum, 40:1879-1885.

Majeed

, El-Khateeb

, El-Shanti

et al. 2005. The spectrum of familial Mediterranean fever gene mutations in Arabs: report of a large series. Semin Arthritis Rheum, 34:813-818.

Majeed

, El-Shanti

, Al-Khateeb

et al. 2002. Genotype/phenotype correlations in Arab patients with familial Mediterranian fever. Semin Arthritis Rheum, 31:371-376.

Mimouni

, Magal

, Stoffman

et al. 2000. Familial Mediterranean fever: effects of genotype and ethnicity on inflammatory attacks and amyloidosis. Pediatrics, 105:1-7.

10.

Olgun

, Akman

, Kurt

et al. 2005. MEFV mutations in Familial Mediterranean fever: association of M694V homozygosity with arthritis. Rhematol Int, 25:255-259.

11.

Paut

, Dubuc

, Sportouch

et al. 2000. Phenotype-genotype correlation in 91 patients with familial Mediterranean fever reveals a high frequency of cutaneomucous features. Rhematology, 39:1275-1279.

12.

Pras

. 1998. Familial Mediterranean fever: from the clinical syndrome to the cloning of the pyrine gene. Scand J Rheumatol, 27:92-97.

13.

Sakallioglu

, Gok

, Kalman

et al. 2006. The phenotype-genotype correlations of FMF patients: single center study. Rheumatol Int, 26:638-640.

14.

Shinar

, Livneh

, Langevitz

et al. 2000. Genotype-phenotype assessment of common genotypes among patients with familial Mediterranean fever. J Rheumatol, 27:1703-1707.

15.

Shinawi

, Riva

, Berant

et al. 2000. Familial Mediterranean fever: high gene frequency and heterogeneous disease among an Israeli-Arab population. J Rheumatol, 27:1492-1495.

16.

Touitou

. 2001. The spectrum of familial Mediterranian fever (FMF) mutations. Eur J Hum Genet, 9:473-483.

17.

Tuglular

, Yalcinkaya

, Paydas

et al. 2002. A retrospective analysis for aetiology and clinical findings of 287 secondary amyloidosis cases in Turkey. Nephrol Dial Transplant, 11:2003-2005.

18.

Tunca

, Akar

, Onen

et al. 2005. Turkish FMF Study Group. Familial Mediterranean fever (FMF) in Turkey: results of a nationwide multicenter study. Medicine (Baltimore), 84:1-11.

19.

Ureten

, Gönülalan

, Akbal

et al. 2009. Demographic, clinical and mutational characteristics of Turkish familial Mediterranean fever patients: results of a single center in Central Anatolia. Rheumatol Int 10.1007/s00296-009-1073-6.

20.

Uyanik

, Albayrak

, Uyanik

et al. 2010. Antibodies directed to cyclic citrullinated peptides in familial Mediterranean fever. Rheumatol Int, 30:467-471.

21.

Yalçinkaya

, Akar

, Misirlioglu

. 1998. Familial Mediterranean fever—amyloidosis and the Val726Ala mutation. N Engl J Med, 338:993-994.

22.

Yalçinkaya

, Cakar

, Misirlioğlu

et al. 2000a. Genotype-phenotype correlation in a large group of Turkish patients with familial Mediterranean fever: evidence for mutation-independent amyloidosis. Rheumatology, 39:67-72.

23.

Yalçinkaya

, Tekin

, Cakar

et al. 2000b. Familial Mediterranian fever and systemic amyloidosis in untreated Turkish patients. Q J Med, 93:681-684.

24.

Yilmaz

, Ozen

, Balci

et al. 2001. Mutation frequency of Familial Mediterranean fever and evidence for a high carrier rate in the Turkish population. Eur J Hum Genet, 9:553-555.