Multiple sclerosis (MS) is a chronic demyelinating disorder pathologically characterized by an infiltration of monocytes and T-lymphocytes into the brain parenchyma, destruction of oligodendrocytes, and the loss of myelin. The role of chemokines and chemokine receptors is particularly important in MS, where myelin-destructive inflammation occurs inside the blood–brain barrier and is related to the influx of peripheral proinflammatory T cells into the CNS. Chemokines play a significant role in the migration of monocytes and T cells and also have been implicated in the onset or progression of MS and experimental autoimmune encephalomyelitis (EAE).1,2

CCR5, a seven transmembrane spanning G protein–coupled receptor, is a specific binding site for the CC-chemokines and is expressed on T-helper cells. Several studies have ascribed to CCR5 surface expression levels an important role in HIV-1 entry and pathogenesis,3 and a CCR5 Δ32 mutation (homozygous deletion) almost invariably protects from HIV-1 infection.4,5 The heterozygotes demonstrate a delay in progression of the disease68 lower viral loads and higher CD4+ counts.9 Aberrant production of chemokines has been described in humans and experimental CNS demyelinating lesions.10,11 β-Chemokine receptors were examined in postmortem MS CNS tissue by immunohistochemistry, and an elevated expression of CCR2, CCR3, and CCR5 was noted.1214 HHV6 virus was additionally found in the human brain specimens and a possible association with MS was suggested.1517 Chemokine receptor expression studies showed CCR5 and CXCR3 to have annualized increase in T2 lesion loads, suggesting these chemokines play an important role in the development of new lesions in MS than in the long-term outcome of those lesions.18

The ligands for CCR5 include RANTES (regulated-upon-activation, normal T expressed and secreted), macrophage inflammatory protein-1α (MIP-1α), and MIP-1β. These ligands belong to the group of CC or β-chemokines and are found principally on Th-1 helper cells.19 Earlier studies in HIV infected individuals have shown that T-cells from CCR5 Δ32 heterozygotes express 5- to 10-fold less CCR5 than wild types after stimulation. In addition, CD4+ T-cell clones from two homozygous persons for the CCR5 Δ32 mutation produced approximately 20-fold more RANTES than wild-type clones.20 Analysis of cytokine and cytokine receptor gene expression in MS samples showed predominantly increased levels of several Th-1 molecules (TGF-ss, RANTES, and MIP-1α), although some Th-2 genes (IL-3, IL-5, and IL-6/IL-6R) were found to be up regulated as well.21

Genetic studies indicate MS is likely to be a polygenic disorder, due to multiple gene associations.2224 Some of the genes reported to be associated with MS include interleukin,25,26 CTLA4,27 HLA-DRB1*1501, and apolipoprotein E.28 Early-published reports29,30 of epidemiological studies of MS concluded that the disease was common in persons of Scandinavian descent: an ethnic group that exhibits a high prevalence of CCR5 Δ32 mutation. Among the European white population there is a north to south gradient of prevalence of the CCR5 Δ32 mutant allele, with the allelic frequencies highest in Scandinavia (16%) and lowest in Sardinia (4%), with a mean allelic frequency across the whole of Europe of 9.1%, where as this variant was found in only 1% in individuals of African origin, 0% in Asians, and 9.8% in Caucasians.31,32

Regardless of the degenerative pathway, damaged or dead neurons are a hallmark of many diseases including Alzheimer's, Parkinson's, glaucoma, ischemia, and multiple sclerosis. Molecules such as cytokines, chemokines, reactive nitrogen/ oxygen species, and proteases play an important role in promoting and/or mediating neurodegeneration.33 Based upon the key role played by chemokines in the migration of macrophages and T cells in MS, and the importance of genetic factors associated and the possible viral or nonviral stimulation involved in the pathogenesis of the disease, we investigated the role of the naturally occurring 32-bp deletion (Δ32 allele) in the CCR5 gene in patients with MS.


Study material

DNA was isolated from postmortem brain tissue taken from 132 MS cases of the Human Brain and Spinal Fluid Resource Center at the VA, Los Angeles Health Care Center, Los Angeles, CA. All cases were necropsy-confirmed based on white matter lesions and demyelination. The sample was composed of 47 male and 85 female, non-Hispanic Caucasians. The age (which was the age-at-death) distribution of MS cases had a range of 30–86, with a mean age of 57.2; the age at onset of MS ranged from 18 to 57, with a mean onset age of 34.6. Chronic progressive (CP) and relapsing remitting (RR) MS subtyping was obtained by chart review for 83 (CP = 38; RR = 45) of the subjects, whereas the remaining charts provided insufficient data for accurate sub typing. The control sample consisted of 163 adult college students from a nearby university (78 males and 85 females) from whom blood samples were obtained. All the healthy subjects included for comparison were non-Hispanic Caucasians, age range between 18 and 49, and a mean age of 34.3. Written informed consent was obtained from the control subjects, and the study was approved by the IRB at City of Hope Medical Center.

Genetic analysis

To detect CCR5 Δ32 mutation, genomic DNA was extracted from brain samples and whole blood by standard procedures. PCR-based assay was used to determine the presence of the CCR5 Δ32 deletion. The following oligo-nucleotides were designed to yield a 232-bp product for the wild type: forward, GRL12A* 5′-TGTTTGCGTCTCTCCCAG-3′ and reverse, GRL12B* 5′-CACAGCCCTGTGCCTCTT-3′

PCR was performed using an initial denaturation step at 94°C for 4 minutes, then a second step with a denaturation 94°C for 45 seconds, annealing at 55°C for 45 seconds, and an elongation at 72°C for 45 seconds for 29 cycles with a final elongation step at 72°C for 6 minutes. PCR products were run on 12% polyacrylamide gels at constant 150 V for 2 hours, stained with ethidium bromide, and viewed on UV. The PCR product obtained had a 232-bp for the wild type and a 32-bp deletion leading to 200 bp for mutant allele. The data obtained on CCR5 Δ32 genotypes were subjected to statistical analysis.

Statistical analysis

Alleles and genotypes were counted and their distributions between groups were determined. The Chi-square (χ2) test was used to statistically compare these groups. All statistical data calculations were done with SPSS statistical package for Macintosh (release 6.1.1) (SPSS Inc., Chicago, IL). We applied survival analysis to compare the survival time after MS onset between the CCR5 genotypes. To estimate the survivor function, Kaplan-Meier product-limit was used34 and Log-rank test was used to examine the differences between two survivor curves of two CCR5 genotypes (normal vs. mutant). We also applied Cox proportional hazard model35 to test the multivariate effects including CCR5 genotypic effect, gender, and MS age of onset. These analyses were done in STATA 7.0 (Stata Corp., College Station, TX).


MS subjects and controls were examined for differences in allelic frequencies and genotype prevalence. No significant variation between these two groups (mutant genotype-MS = 19.7%; Control = 17.1%; P = 0.394) or with age of onset of MS (<35 and ≥ 35years; P = 0.572) was found with respect to CCR5 genotypes. However, a marked difference was observed when the Chi-square test was used to examine the potential progressive increase in certain genotypes of MS subjects across five groups with years of survival from clinical onset to death: ≤5 years, 6–10 years, 11–15 years, 15–20 years, and >21 years (see Table 1). The MS subjects with 11 genotype (normal) survived progressively more years compared to subjects with 12 and 22 genotypes (single or double mutant alleles) (P = 0.00001).

Table 1 Frequency of CCR5 Δ32 genotypes in MS patients in relation to years of survival (age at onset to death)

Additionally, these groups were examined for possible differences in mutant allelic frequencies by years of survival. A significant association of the CCR5 mutant allele 2 with an early death is shown in Figure 1 with a progressive reduction in the years of survival from age at onset to the age of death with increased frequency of the CCR5 deletion allele (PL = 0.00005). The patients carrying the CCR5 normal allele 1 had an earlier age at onset (mean 33.96 years) and longer survival years (mean 25.48 years) as compared to patients carrying at least one CCR5 mutant allele 2 (mean onset 35.36 years; mean survival 17.04 years).

Fig. 1
figure 1

Frequency of the CCR5 Δ32 mutant allele in MS patients in relation to years of survival (age of onset to death).

The overall median survival time is 23 years (range 1–56) after the onset of MS. The median survival times are 24 and 16 years and death rates are 0.039 and 0.056 for MS patients with genotypes 11 (normal) and 12 + 22 (mutant), respectively. Figure 2 shows Kaplan-Meier survival curves for patients with CCR5 genotypes. The cumulative death probability is significantly higher (P = 0.00099) in the mutant (12 + 22 genotypes) group than in normal (11 genotype) group. The cumulative death probabilities are 34% for the 11 genotype group and 69% for the 12 + 22 genotype group 20 years after onset. The genotype was marginally significant with clinical subtype (P = 0.033). Because this difference can be confounded by clinical MS subtype (mutant allele in RR = 27.9% and CP = 7.9%) and the age of onset, we used the Cox regression model to test the main effect of CCR5 with simultaneous adjustment with these potential effects. The results showed that MS subtype was highly significant (P = 0.0001) with a greater tendency for the RR type, whereas gender and age of onset was not significant. The hazard ratio was 2.12 for 12 + 22 (mutant) versus 11 (normal) genotype with P = 0.006 (Table 2), after adjusting for MS clinical subtype (CP and RR type). This means that MS patients with CCR5 mutant genotype (12 and 22) have over twice the mortality as compared to the normal genotype,11 even after adjusting for the subtype. We have included sex as a covariate in the model and the results did not change, suggesting CCR5 and MS subtype to be the two strong covariates of survival.

Fig. 2
figure 2

Kaplan Meier product-limit survival curves for MS patients separated by CCR5 11 and 12 genotypes.

Table 2 Effect of CCR5 Δ32 on survivorship in MS patients by cox proportional hazard model, with adjustment of MS subtype (N = 83)

Considering the high frequency of females with MS we believed that there could be a different mechanism in males versus females. Thus, we stratified our sample to separately investigate this by gender. Interestingly, this genotype effect was primarily associated with female patients (Table 3). The hazard ratio in females was 3.58 (P = 0.001; 95% CI = 1.74–7.38). There is a potential bias in as far as those patients with unknown subtypes may have worse clinical diagnosis and shorter survival years. However, this effect more likely forces our estimate downward (i.e., more conservative) if these patients were associated with CCR5 mutant genotypes.

Table 3 Effect of CCR5 Δ32 on survivorship in MS patients by cox proportional hazard model, with adjustment of MS subtype (N = 50)


T-cell chemokine expressions in human autoimmune disorders are believed to arise from inappropriate Th1- and Th2-dominated responses. Th1 cells express CXCR3 and CCR5 and Th2 cells express CCR3 and CCR4. Chemokine receptors expressed by Th1 cells may be important as reports imply an association of MS with CNS infiltrates of CD4 cells and monocytes. To date, studies on survival of MS in relation to genotypes have not been specifically addressed and causes of death in MS cases have not been well studied. In a study of 6,068 MS cases from the Danish Multiple Sclerosis Registry, MS was noted on the death certificate as the underlying cause of death in 55.4% of cases,36 but the mechanism by which MS contributes to “premature” death has not been explicated. However, young age at onset and an initial remitting clinical course have both been shown to be significantly associated with longer survival.37 Previous studies have examined the role of the CCR5 polymorphism and MS. One study showed that the presence of a CCR5 Δ32 mutation resulted in a 3.2-year delay in the onset of familial MS,38 and the other showed that the age of disease onset was somewhat lower (about 3 years) in patients carrying the CCR5 Δ32.39 In contrast to these two studies, another report failed to demonstrate any difference.40 Although these reports have demonstrated later age at onset and increased disease intervals in CCR5 Δ32 mutants compared to wild-type alleles, these observations were not on postmortem samples and did not predict for overall survival. On the basis of the suggested links between cytokines and MS, and for the reasons given earlier, we sought to examine the possible association between CCR5 and MS.

Our study on postmortem samples failed to show an association of CCR5 deletion mutation with MS as compared to healthy controls, but showed a significant gradient in survival for MS patients when individuals having mutant CCR5 genotypes were compared to those with wild type. In diseases with multiple gene effects, genes not involved in disease susceptibility may still determine severity acting alone or could become susceptible candidates in combination with alleles of other genes. We observed that patients carrying at least one CCR5 mutant allele had an estimated 1.8 years delayed onset and 8.44 years earlier death. These observations only imply that those individuals with CCR5 mutation are predisposed to earlier mortality. An earlier report37 on association of young onset MS cases with longer survival supports our results. However, the contrast in results observed in other studies may be due in part to limitations imposed by sample size in each study, and also the reduced information on survival.

Chemokine receptor expression on T cells is related to new lesion development in MS,18 and the one possible explanation for the strong association of mutant CCR5 allele with early death observed might be the result of enhanced CNS damage. An alternative explanation could be given based upon the CCR5 Δ32 mutation causing an up regulation of RANTES and directing the immune response away from the Th-1 to Th-2 cells. The overall main effect of CCR5 gene observed with significantly higher cumulative death probability in mutant genotype group, in MS subtypes and in females (Table 2 and 3) suggests a complex regulation of the inflammatory response and the possible association of chemokine receptor with disease activity.

Relevant cytokine interactions are expected to vary depending on the disease process. A dysregulation of the cytokine balance can generate deleterious effects resulting in neurotoxicity, neurodegeneration, increased pathogenicity, and neurological complications. Thus genes encoding various cytokines are logical candidates to assess possible associations with MS susceptibility and the course of the disease. Our studies on postmortem samples imply that those individuals with CCR5 mutation are predisposed to earlier mortality. Thus, the presence of CCR5 mutation in patients with MS showed an exactly opposite survival effect as compared to the protective effect seen in patients with HIV-1 infection, i.e., carriers of this mutant allele were either resistant to infection by HIV-1 or had slower progression of AIDS41,42 unlike MS patients who manifested earlier death.

In polygenic disorders, genes not involved in disease susceptibility per se may play a role in determining survival. The predominance in mortality of females over males may also suggest a role for estrogenic hormones in this pathogenic milieu. In an earlier study a significant association between an X-linked gene GABRA3 and MS was observed.43 Combined, these findings are consistent with the theory that generation of clinical manifestations involves both cytokine action and cytokine modulation of neurotransmitters. Understanding the role of the CCR5 gene and characterization of cytokine-cytokine interaction and cytokine-neurotransmitter interactions is important. The clinical relevance of this CCR5 mutation in patients with MS could help in designing a therapeutic regimen and may serve as a prognostic marker.