Deficiency of mannan-binding lectin associated serine protease-2 due to missense polymorphisms


Mannan-binding lectin (MBL) and ficolins distinguish between self, non-self and altered-self by recognizing patterns of ligands on the surface of microorganisms or aberrant cells. When this happens MBL-associated serine protease-2 (MASP-2) is activated and cleaves complement factors to start inflammatory actions. We examined human populations for MASP-2 levels, MASP-2 function and for the presence of mutations in coding exons of MASP2. The MASP-2 levels were lowest in Africans from Zambia (median, 196 ng/ml) followed by Hong Kong Chinese (262 ng/ml), Brazilian Amerindians (290 ng/ml) and Danish Caucasians (416 ng/ml). In the Chinese population, we uncovered a novel four amino-acid tandem duplication (p.156_159dupCHNH) associated with low levels of MASP-2. The frequency of this mutation as well as the SNPs p.R99C, p.R118C, p.D120G, p.P126L and p.V377A were analyzed. The p.156_159dupCHNH was only found in Chinese (gene frequency 0.26%) and p.D120G was found only in Caucasians and Inuits from West-Greenland. The p.P126L and p.R99Q were present in Africans and Amerindians only, except for p.R99Q in one Caucasian. The MASP-2 levels were reduced in individuals with p.V377A present. The MASP-2 present in individuals homozygous for p.377A or p.99Q had a normal enzyme activity whereas MASP-2 in individuals homozygous for p.126L was non-functional.


The immune system has evolved under the selection pressure of being able to eliminate infections while at the same time avoiding injuring the host. A number of measures are needed to keep the system in balance. The innate immune system relies on a balance of single nucleotide polymorphisms (SNPs) in the genes encoding the many members of this system.1, 2 The recognition of non-self structures (e.g., microorganisms) by the innate immune system is based on the recognition of certain patterns of ligands (pathogen associated molecular patterns) by cell-bound or soluble pattern recognition molecules (PRMs).3 Mannan-binding lectin (MBL) and the ficolins (L-ficolin, M-ficolin and H-ficolin) are examples of soluble PRMs.4, 5 In the case of MBL, the recognized motifs are patterns formed by common carbohydrates.4, 6 In the case of ficolins, the recognized pattern may be formed by N-acetyl groups as presented by acetylated sugar or amino acids.7, 8, 9 The binding of MBL or ficolin to a microorganism will initiate the activation of the complement system. This occurs through an activation of one or more of three MBL-associated serine proteases (MASPs): MASP-1, MASP-2 and MASP-3.10, 11, 12, 13

The complement system may be activated by three pathways, the classical (when the C1 complex is activated), the alternative (when spontaneously formed C3b is deposited on a non-inhibitory surface) and the lectin pathway (when either the MBL/MASP complex or a ficolin/MASP complex is activated).14, 15 When this occurs a number of anti-microbial activities are initiated, but at the same time factors potentially harmful to the body are produced. It is thus important to study the influence of the genetic background on the delicate balance of levels and activities of the proteins of the complement system.

A number of polymorphisms in the MBL encoding gene, MBL2, influence the level of MBL.4, 16 Low MBL levels have been associated with an increased susceptibility to infections in various situations.17, 18, 19 We have previously identified a mutation in MASP2, the gene encoding MASP-2, resulting in a missense (also called a non-synonymous mutation, denoting a mutation resulting in a codon for a new amino acid) SNP, c.359A>G (where bp 359 corresponds to the cDNA position, counting from the methionine starting codon) (Table 1).20 This leads to an exchange of aspartic acid for glycine at residue 120 (D120G, when numbering according to the synthesized protein, that is, including the leader sequence) (Table 1). The presence of this SNP lowers the MASP-2 concentration and the MASP-2 in one patient homozygous for p.120G was found to lack the ability to activate complement.20 Four other MASP2 missense SNPs have been reported in the literature, that is, c.251G>A (p.R99Q), c.307C>T (p.R118C) and c.332C>T (p.P126L)21 and c.1130T>C (p.V377A),22 and another three missense SNPs, c.464A>G (p.H155R), c.1111G>T (p.D371Y) and c.1316G>A (p.R439H) are found in the NCBI SNP database (Table 1). None of these have been analyzed regarding the influence on levels or functions of MASP-2.

Table 1 Nomenclature of the MASP2 polymorphisms mentioned in the text

The frequencies of the p.R99Q, p.R118C, p.D120G and p.P126L SNPs were previously analyzed only in Danes and Spaniards. We now determine the frequency in other populations, Hong Kong Chinese, Zambian Africans, Amerindians in Brazil and Inuits from Greenland. We have further sequenced the exons of MASP2 among Chinese and Africans with low MASP-2 levels and have found a new polymorphism. We have analyzed the activity and levels of MASP-2 in individuals with the different MASP2 genotypes in various populations.


MASP-2 levels in different ethnic populations

We measured the levels of MASP-2 in plasma from individuals from various ethnic groups. The levels were lowest in Africans (median 196 ng/ml, range 8–926) followed by Hong Kong Chinese (262 ng/ml, 52–720), Amerindian (290 ng/ml, 35–1300) and Caucasians (416 ng/ml, 125–1152) (Figure 1). There is a significant difference between all groups (P<0.001). The difference between single groups, for example, Africans and Caucasians (P<0.0001) are highly significant.

Figure 1

The MASP-2 levels in different ethnic populations, Caucasian Danes (n=350), Hong Kong Chinese (n=200), African Zambians (n=194) and Amerindian Brazilians (n=324). Bars represent medians and boxes indicate IQRs, whiskers show 10th and 90th percentiles and outliers are indicated.

We have previously suggested that a MASP-2 level below 100 ng/ml would be suggestive of MASP-2 deficiency as no Caucasians below this level was found in an earlier investigation,23 except in an individual homozygous for p.D120G. Now we find levels below 100 ng MASP-2/ml in 35 (18%) of the 200 Zambians. This is a significantly higher frequency of low MASP-2 concentration than seen in the Caucasian (χ2 test is allowed, P<0.0001). There is no difference between the Chinese and the Amerindians (P=0.71) whereas there are significantly fewer Caucasians than Chinese and Amerindians with levels below 100 ng/ml (P=0.01).

A 12 bp tandem duplication

The MASP-2 concentration in the plasma from 200 Hong Kong Chinese individuals was measured. The nucleotide sequences of the exons of MASP2 were determined in the four individuals with the lowest levels (52, 59, 71 and 89 ng MASP-2 per ml, respectively). In one person the duplication, c.466_477dupTGCCACAACCAC, was identified (Table 1). This should lead to a tandem duplication of four amino acids, 156_159dupCHNH, in the EGF domain of MASP-2 (Figure 2). We now analyzed for the presence of this tandem duplication in genomic DNA from a total of 573 Hong Kong Chinese. Three persons were found to be heterozygous for the duplication (Table 2). The presence of the duplication is associated with low levels of MASP-2 as the three heterozygous presented MASP-2 levels below 113 ng/ml (113, 100 and 52 ng/ml, respectively) as compared to the median value of 264 ng/ml (range 59–720 ng/ml) for the wild-type individuals (P=0.004 by rank sum test).

Figure 2

The positions of the amino acid exchanges and the tandem duplication in MASP-2. MASP-2 is composed of an N-terminal CUB (C1r/C1s, Uegf and bone morphogenetic protein-1) domain, followed by an EGF (epidermal growth factor-like), a second CUB domain, two CCP (complement control protein) domains and a serine protease domain.11, 13 The numbering is referring to the protein including the leader sequence of 15 amino acids. The lengths of the domains are drawn proportional to the number of amino acids in the domains. The polymorphisms of the indicated amino acids changes p.H155R, p.D371Y and p.R439H are listed in the NCBI refSNP database (Table 1), but are not addressed in the present report.

Table 2 The frequency of MASP2 alleles in different populations

The calculated frequency of this allele is 0.26% (3/1146) (Table 2), that is, an expected occurrence of homozygous individuals in approximately seven out of 1 × 106. We tested for the presence of the tandem duplication in individuals from four other populations, Danish Caucasians (n=350), Zambian Africans (n=194), Amerindians from Brazil (n=324) and Inuits from Greenland (41 from West Greenland, 96 from East Greenland). We did not find the duplication to be present in any of these populations (Table 2).

Although the duplication was associated with very low levels of MASP-2, the three heterozygous individuals showed MBL pathway activity comparable to that measured in individuals without the mutation and having comparable MBL levels (not shown). This follows earlier observations that there is no correlation between the level of MASP-2 protein and the activity of the MBL pathway (see below). It appears that only in individuals homozygous for mutated MASP-2 may the ability to activate the MBL pathway be wrecked.20

The p.D120G polymorphism

We have previously described the presence of the SNP c.359A>G in a Caucasian population, leading to the amino acid exchange p.D120G (Table 1 and Figure 2). We tested Chinese, Caucasian, African, Amerindian and Inuit individuals for the presence of this SNP and found it only in the Caucasian population (27 heterozygous out of 350, that is, a gene frequency of 3.9%, Table 2), and in individuals from West Greenland (3 out of 41 individuals were heterozygous, gene frequency of 3.7%). In the Caucasian cohort a median value of 427 ng/ml (IQR 351–524) and 224 ng/ml (IQR 183–257) were determined in wild type and heterozygous individuals, respectively (in this calculation, we omitted eight samples with a different genotypes – see later). The levels in the two groups differ significantly (Wilcoxon rank sum test, P<0.0001).

The polymorphisms p.R99Q, p.R118C, p.P126L and p.V377A

The exons of MASP2 in three Zambians with very low MASP-2 levels (sample 26:37 ng/ml, sample 138:48 ng/ml and sample 198:85 ng/ml) and with deficiency in the ability to deposit C4 on mannan (sample 26:30 mU/ml, sample 138:34 mU/ml and sample 198:42 mU/ml) were sequenced. Two of the samples (sample 26 and 138) had normal MBL level whereas one (sample 198) had low MBL level. The latter sample was still deficient in MBL pathway activity (46 mU/ml) after addition of recombinant MBL to 5 μg/ml.

The sequences obtained revealed two SNPs:c.296G>A (p.R99Q) and c.377C>T (p.P126L) (Table 1). These two as well as SNP, c.352C>T (p.R118C), were reported previously,21 and a fourth SNP, c.1130T>C (p.V377A) was reported by Stover et al.22 (Table 1). The presence of the alleles described above was analyzed in the samples from the different populations (350 Danish Caucasians, 573 Hong Kong Chinese, 194 Zambian Africans, 324 Brazilian Amerindians and 137 Greenlandic Inuits). The frequencies found are given in Table 2. Owing to lack of DNA not all of the SNPs were tested in the Hong Kong samples.

One Caucasian individual and two Amerindians were found to be heterozygote (p.99R/Q) with regards to the SNP p.R99Q, as compared to 30 found among the 194 Africans who also presented one homozygous individual, whereas this mutation was not present in the Chinese or the Inuit samples. The SNP p.P126L was present at a high frequency in Zambians whereas it was absent in the other populations. The SNP p.R118C was not found in any of the studied individuals. The SNP p.V377A was present in a high frequency (17%) in Africans and was also found in Caucasians (Table 2).

To get an indication of possible haplotypes we analyzed the African data using the program Haploview (version 3.32, using four gamete rule), which revealed four haplotypes [99R; 126P; 377V], [99R; 126P; 377A], [99R; 126L; 377V] and [99Q; 126P; 377V] at frequencies of 0.59, 0.17, 0.16 and 0.08, respectively.

In Figure 3a the MASP-2 level in black Zambians of different genotypes is depicted. Of the 194 tested individuals, 68 (35%) were homozygous for the haplotype [99R; 126P/P; 377V] (median MASP-2 concentration of 241 ng/ml, IQR 169–363), the heterozygous genotype [99R; 126P; 377V]/[99R; 126P; 377A] (median MASP-2 concentration of 170 ng/ml, IQR 126–252) was found in 38 (20%) individuals and eight (4%) individuals were homozygous for [99R; 126P; 377A] (median MASP-2 concentration of 89 ng/ml, IQR 54–142). Heterozygocity for the haplotypes [99R; 126P; 377V] and [99Q; 126P; 377V] (median MASP-2 concentration of 361 ng/ml, IQR 288–485) was found in 21 (11%) individuals, two (1%) were heterozygous [99R; 126P; 377A]/[99Q; 126P; 377V] (MASP-2 concentrations 214 and 282 ng/ml), and 1 (0.5%) was homozygous for [99Q; 126P; 377V] (MASP-2 concentration 611 ng/ml). In 35 (18%) individuals the genotype [99R; 126P; 377V]/[99R; 126L; 377V] (median MASP-2 concentration of 163 ng/ml, IQR 101–225) was found, the genotype [99R; 126P; 377A]/[99R; 126L; 377V] (median MASP-2 concentration of 125 ng/ml, IQR 67–152) was found in eight (4%) individuals, the genotype [99R; 126L; 377V]/[99Q; 126P; 377V] (MASP-2 concentration; median 120 ng/ml, IQR 86–159) was found in seven (4%) individuals and six (3%) individuals were found to be homozygous for the haplotype [99R; 126L; 377V] (median MASP-2 concentration of 88 ng/ml, IQR 24–88).

Figure 3

MASP-2 concentrations in individuals of different genotypes. (a) Zambian Africans and (b) Danish Caucasian. Within each ethnic group the concentration for each individual is given for the genotypes found. Within each genotype the amino acids in front of the slash belongs to the same suggested haplotype (e.g. 99R; 126P; 377V for the first group on the left) as does the amino acids after the slash (also 99R; 126P; 377V in the first group). As mentioned in the text and indicated here in this figure four haplotypes for the Africans are proposed. The Africans are all p.[118R/R; 120D/D] and have no dupCHNH. The Caucasians are all p.[118R/R; 126P/P] and have no dupCHNH. The median is indicated where relevant as a horizontal line in the different genotypes; the actual numbers are given in the text. The number of individuals in each group is given below the x-axis.

The data reveals that there is a significant difference in MASP-2 levels between the group homozygous for [99R; 126P; 377V] (median 241 ng/ml) and the heterozygous group [99R; 126P; 377V]/[99R; 126P; 377A] (median 170 ng/ml) (P=0.003) and the difference is obvious (Figure 3a) between these two groups and the group homozygous for [99R; 126P; 377A] (median 89 ng/ml). However, since only eight individuals were homozygous no meaningful statistics can be performed on this. The presence of p.377A in the African population thus leads to lower MASP-2 levels. The influence of p.V377A on the MASP-2 level in Caucasians is illustrated in Figure 3b. The MASP-2 concentrations in Caucasians with the heterozygous genotype [99R; 120D; 377V]/[99R; 120D; 377A] have a median of 384 ng/ml (IQR 330–414) as compared to the group homozygous for [99R; 120D; 377V] where a median value of 428 (IQR 351–524) is seen. The one Amerindian homozygous for p.377A has a MASP-2 level of 199 ng/ml as compared to the median of 290 ng/ml in all the Amerindians studied.

There may be a negative influence on the MASP-2 level in Africans of the p.P126L allotype since there is a difference between the group being homozygous [99R; 126P; 377V] and the group of heterozygots [99R, 126P; 377V]/[99R, 126L; 377V] (P=0.002) (Figure 3a). The number (six) of homozygous for the haplotype [99R, 126L; 377V] is too small for statistic analysis, but the median in this group is 88 ng/ml, as compared to 163 ng/ml in the heterozygote and 241 ng/ml in the wild type individuals.

In Africans the p.R99Q allotype seems to be associated with increased levels of MASP-2 as a difference is seen between the individuals homozygous [99R; 126P; 377V] and the heterozygous group [99R; 126P; 377V]/[99Q; 126P; 377V] (P=0.007), and the one individual homozygous for [99Q; 126P; 377V] has a high level of 611 ng MASP/ml. The one Caucasian being heterozygous p.99R/Q did show a MASP-2 level in the high end (Figure 3b).

MBL pathway activity

We further studied the ability of plasma from individuals of various genotypes to deposit C4b on mannan-coated microtiter wells, that is, the MASP-2 activity of the MBL pathway of complement activation. The ficolins do not bind to mannan, but it is assumed that MASP-2 mutations will affect MBL and ficolin-mediated complement activation similarly, as was shown for the p.D120G mutation.13

The MASP-2 present in the African individuals homozygous for p.377A (all p.[99R/R; 126P/P; 377A/A]) had an activity similar to the ones homozygous for p.377V MASP-2 (e.g., p.[99R/R; 126P/P; 377V/V] (Figure 4a). The MBL pathway was also normal in the one individual homozygous for p.99Q (Figure 4a).

Figure 4

MBL pathway activity. (a) The MBL pathway activity (C4b deposition capacity, mU/ml) in Zambians versus the MBL levels (r=0.32). Indicated with open triangles are the three samples from which the sequences of MASP2 exons were obtained. The eight open squares are individuals homozygous for p.377A. The six open diamonds are individuals homozygous for p.126L. The round circle is an individual homozygous for p.99Q. (b) The MBL pathway activity of Zambian Africans versus MASP-2 levels. (c) The MBL pathway activity of Danish Caucasians versus MBL levels (r=0.85). (d) The MBL pathway activity of Danish Caucasians versus MASP-2 levels.

On the other hand, the MASP-2 of the individuals homozygous for p.126L had very low C4b deposition activity (five were below 100 mU/ml and the last below 150 mU/ml). Of special interest is that even the plasma with a MBL concentration near the normal median level, 1436 and 811 ng MBL/ml, were low in activity, 86 and 70 mU/ml, respectively.

When comparing the correlation between C4b deposition and MBL levels (Figure 4a and c) or between C4 deposition and MASP-2 levels (Figure 4b and d) it is seen that the C4b deposition is strongly associated with the MBL level but not with the MASP-2 level. The individuals in, for example, Figure 4d, with rather high MASP-2 levels and low C4 deposition all presented low MBL levels. The association between C4 deposition and MBL levels is more pronounced in the Caucasian population (r=0.87) (Figure 4c) than in the African population (r=0.323) (Figure 4a), reflecting the higher prevalence of genotypes associated with low levels of MASP-2 and non-functional MASP-2 in Africans (Figure 4b and d).


Inflammation is a key element in the response of the innate immune system to a variety of challenges, including those provided by bacterial and viral infection as well as by damaged or dying host cells. When the normal regulatory mechanisms of the complement system are disturbed, the potential for developing chronic inflammatory diseases is increased.14, 15 In the present report, we have studied polymorphisms in MASP2 and their effect on the levels and function of MASP-2, an important molecule in the activation of the complement system via MBL and ficolins.4, 13

We previously reported on a MASP-2 mutation, p.D120G, associated with low MASP-2 level and functional deficiency in Caucasians,20 and we wished to extend this line of investigation to other ethnic groups. We find 7.7% of a new group of 350 Danish Caucasians to be heterozygous for the SNP c.359A>G leading to the p.D120G exchange, indicating a gene frequency of 3.9%. Carlsson et al.24 found in a Swedish population 14 heterozygotes among 112 patients with cystic fibrosis (CF) (a gene frequency of 6.3%), and five heterozygotes among 200 healthy individuals (a gene frequency of 1.3%). In a study of patients with psoriasis and family members, 894 individuals were tested for p.D120G and a total of 62 heterozygotes and one homozygote were found, giving a gene frequency of 3.6% (the allele was not associated with psoriasis).25 Homozygocity was reported in one individual in a group of 293 Polish children with respiratory infections26 and in one child with CF.27 We have encountered further three homozygous individuals, two admitted to the lung clinic at Aarhus University Hospital, and one colon cancer patient (not published). A new study of a Spanish population found two homozygous out of 2008 individuals (including 967 pneumonia patients, 130 SLE patients, 43 children with recurrent respiratory infections and 868 healthy individuals).28 No disease association was indicated with the variant allele and the two homozygous persons were found among the healthy individuals. One must thus conclude that as for MBL, and indeed for other complement components as well, deficiency in itself does not result in disease, rather, it is a modifier, which may penetrate when also other elements are compromised. We here confirm that in the Caucasians the D120G substitution is the primary cause of lower MASP-2 levels (Figure 3b).

We did not find the p.D120G allotype in the non-Caucasians tested except for the presence in Inuits from West-Greenland (Table 2). The Inuits from East Greenland did not present this polymorphism. They are believed to be an ethnically more homogeneous population than the West Greenland Inuit population, where there is more admixture of Caucasian genes. While HLA typing of the West Greenlanders of the present study identified them as Inuits,29 admixture of Caucasians genes can not be excluded, and the small sample does not allow for strong conclusions. Nevertheless it is tempting to speculate that the finding might reflect true ethnic differences, connected to the population of Greenland taking place in several migration waves separated by thousand of years, the last, by the Thule groups, only one thousand years ago.30 The absence of the p.D120G allele in Chinese is supported by a report examining the influences of MBL2 and MASP2 genotypes on susceptibility to severe acquired respiratory syndrome (SARS). In all 1757 Asian individuals tested no p.D120G allele were found.31

Individuals homozygous for the p.D120G mutation have no MBL pathway activity.20 Except for this situation we surprisingly find no correlation between MASP-2 concentrations and MBL pathway activity. We interpret this as indicating that MASP-2 is present in excess. It follows that by far the most cases of deficiency to deposit fragments of complement factors C4 on a mannan surface is caused by MBL deficiency, as indeed was the original finding when the influence of MBL on opsonic activity was discovered.17 This is illustrated in Figure 4 where a close association is seen between MBL concentration and C4b depositions. This correlation is clearly more pronounced in the Danish than in the Zambian individuals (Figure 4a and c) (see below for a discussion of the higher frequency of MASP-2 deficiency in Africans). Carlsson et al.24 reported an association between complement factor C9 deposition on mannan and heterozygocity for p.D120G, but this could only be detected if individuals with low MBL were excluded from the analysis, and only for CF patients, and not for healthy individuals. It is important to note that when assaying for C4b deposition, the activity seen is due to MASP-2 only, whereas the deposition of C9 is dependent on a range of other complement factors, including control proteins. As seen in Figure 4b and d there is no obvious association between MASP-2 levels in general and C4 deposition.

Owing to lack of functional assays we do not know if the MASP-2 levels influence the activity of the other complement-initiating pathogen recognition molecules of the lectin pathway, which are also known to bind MASP-2, that is, activation via L-, H- or M-ficolin. However, we do know that recombinant MASP-2 of the p.120G type binds to none of the ficolins.13

Sequencing the exons of MASP2 in Hong Kong Chinese individuals who were low in MASP-2, revealed a polymorphism involving the duplication of 12 bp, resulting in the insertion of four amino acids (p.156_159dupCHNH). This duplication was only found in Chinese and not in any of the 1005 non-Asians tested (Table 2). Owing to the low frequency of the duplication in Chinese these numbers are too small to state that the duplication is not present in Caucasians (P=0.29), Africans P=0.58), Amerindians or Inuits. The MBL pathway activity was not affected in individuals heterozygous for the duplication, likely due to activity of the MASP-2 produced from the non-mutated gene. We speculate that MASP-2 in individuals homozygous for this polymorphism may be non-functional, that is if any MASP-2 at all is produced.

Sequencing the MASP2 exons of three Africans with low MASP-2 levels as well as low MBL pathway activity revealed a SNP, c.377C>T, giving rise to a P126L exchange and a SNP, c.296G>A, giving rise to a R99Q exchange. The p.R99Q SNP was previously identified in 4 out of 15 individuals from Sub-Saharan origin and p.P126L in 4 out of 50 North Africans, presumably of Arabic ethnicity.21 Lozano et al.21 further reported the presence of a p.R118C (c.352C>T) polymorphism in 2 out of the 50 North Africans. We did not detect the p.R118C in any of the 194 Zambians blood donors nor in any of the other populations we analyzed. This missense mutation could be a characteristic of North African Arabics or could be a familial trait (no information on family relations were given).

The p.V377A mutation was reported previously by Stover et al.22 A gene frequency of 0.882 for c.1130T (p.377V) and 0.118 for c.1130C (p.377A) is given in the NCBI SNP database, which is similar to the 0.83/0.17 we found in the Africans, whereas in contrast we found 0.99/0.01 in Caucasians and 0.93/0.07 in Inuits from West Greenland and 0.96/0.04 in Amerindians (P<0.01 between Africans and the other groups).

When studying Danish Caucasians, a MASP-2 level below 100 ng/ml indicated MASP-2 deficiency as only individuals homozygous for the p.D120G mutation show a MASP-2 level below this (Moller-Kristensen et al.23 and unpublished data). We now find that this interpretation does not apply when studying other populations. Thus, despite the absence of the p.120G mutation, of the 324 Amerindians tested eight (2%) were below 100 ng/ml and also a number of the Hong Kong Chinese and Zambian Africans presented MASP-2 levels below 100 ng/ml, whereas we still did not see any below this level in the healthy Danish Caucasians (Figure 1).

Of the 200 Chinese tested nine (5%) were below 100 ng/ml. Only one of these had the p.156_159dupCHNH and none of the other two mutations tested for were found. We have focused on exon sequences in the present report. Possible explanation for the low levels could be polymorphisms in the promoter region or intron regions of MASP2. The promoter region of human MASP2 has been described.22

MASP-2 levels below 100 ng/ml were much more common in Africans (19%, 38 out of the 200 tested) than in the other populations. A possible explanation for this could be the unique presence of the SNPs p.P126L and p.R99Q in the African population. However, we cannot rule out (due to the number of individuals tested in other populations) that these allotypes are simply found in higher frequency in Africans. The individuals homozygous for p.126L (the haplotype [99R; 126L; 377V]) are all low in MASP-2 (Figure 3a). On the other hand, the presence of p.R99Q (the haplotype [99Q; 126P; 377V] is associated with elevated MASP-2 levels (Figure 3a).

The allele p.377A/V influences the MASP-2 level. Among the Africans eight were homozygous for p.377A (haplotype [99R; 126P; 377A]), all of them with low MASP-2 levels (Figure 3a), and the heterozygous individuals were also lower than wild type individuals (Figure 3a). The high frequency of the allele p.V377A in Africans will drive this population towards lower MASP-2 levels. This may point to the importance of a delicate balance between the anti-microbial and the potentially harmful activities of the complement system. A parallel hypothesis has been proposed to explain the differences in frequencies of polymorphisms in MBL232 as well as for other genes of the immune system.33

There are missense polymorphisms listed in the NCBI refSNP database, which we did not test for (Table 1 and Figure 2). These were not studied in the present report as we did not find them in any of the individuals with low MASP-2 levels, which were analyzed. The c.464A>G (p.H155R) is reported to have an average allele frequency of 0.998 for c.464A and 0.002 for c.464G, the c.1111G>T (p.D371Y) an average allele frequency of 0.452 for p.371G and 0.548 for p.371T, whereas for c.1316G>A (p.R439H) no frequency is given.

The available structural information on MASP-2 may yield clues as to the possible functional effects of the various mutations or allotypes. Two MASP-2 molecules form a dimer, which is suggested to be the form binding to MBL and ficolins.34, 35 MAp19, which also binds to MBL and the ficolins, is encoded by exons b, c, d and e of MASP2,22 that is, MAp19 and MASP-2 contain identical CUB1-EGF domains, but MAp19 has four unique amino acids added to the C-terminal end (encoded by exon e).36 MAp19 has no known biological activity. We have no specific tests for MAp19, and thus we do not know if the reported polymorphisms influence the concentrations or functions of this protein.

Considering the missense polymorphisms in the CUB1 domain the p.D120G mutation has a known functional effect, that is, the mutated MASP-2 cannot bind to MBL or to ficolins.13 A charged residue at this position in the CUB domain act as part of a calcium-binding site.37 The other substitutions in the CUB1 domain (p.R99Q, p.R118C and p.P126L) are clustered near the amino acids suggested to be involved in the binding of MASP-2 to MBL or the ficolins (Y74, E98, Y121 and E124, numbering including the leader sequence).37 The p.R99Q SNP is likely to be of little impact since glutamine is found at the homologous position in the MASP-1 sequence (Q100 in MASP-1, numbering including leader peptide) and as this residue is oriented in the direction opposite to the calcium-binding site. In accordance with this the one person homozygous for p.99Q had an active MBL pathway (Figure 4a). The R118 is closer to the binding site and a charge replacement (R118C) at this position may lead to functional effects. As stated above, we do not find this mutation in any of the individuals tested (Table 2). The p.P126L may have an effect through changing the orientation of the nearby E124. We suggest that this may be the reason for the non-functional MASP-2 seen in the individuals homozygous for p.126L (Figure 4a).

The CHNH sequence (aa 156–59) (Figure 2) is highly conserved in EGF domains (especially the CHN156-158), and provides part of the calcium-binding group in the MASP-2 EGF domain.37 Introducing an insertion into this place is likely to alter this binding site and thus the dimer interface with functional implications. The other reported substitution (p.H155R) in the EGF domain is at a position that belongs to the dimerization interface. However, the substitution of histidine to arginine is likely to have only minor effects.

The two substitutions D371Y and V377A, are encoding amino acids in the CCP2 domain (Figure 2). D371 is a surface residue surrounded by numerous aromatic residues. From the crystallographic information38 it appears that changing the aspartic acid to tyrosine can be accommodated without disturbing the overall structure. V377 is directed towards the inner core, thus the mutation of this valine to an alanine could introduce a decrease in the hydrophobic interaction, as suggested by the program SNPs3D ( Anyway, we did not see any functional effect when testing homozygous (p.377A/A) individuals (Figure 4a), although we did see a lowering effect on the MASP-2 level (Figure 3a).

One reported substitution (p.R439H) is located in the link region (also referred to as the activation peptide) (Figure 2). R439 corresponds to the P6 position of the cleavage site between the A and the B chain of MASP-2 (P1 being the primary cleavage site). It corresponds to the major polar region 3 proposed by Gal et al.38 to be important in the activation process. Although a change to histidine may not have dramatic structural consequences, a functional impact cannot be excluded.

In future analysis of complement in different patient groups, we suggest that MASP2 polymorphisms may be included together with the well-established analysis for polymorphisms for MBL2. At the protein level expression of recombinant MASP-2 or MAp19 with the amino-acid exchanges described in the present report should be analyzed for synthesis efficiency and functional activity, that is, binding to MBL and ficolins and activation of other complement factors. Such studies could lead to a description of combined polymorphisms, which might possibly lead to higher disease susceptibility or alternatively to less harmful reactions against self structures.

Materials and methods

Plasma and DNA

Blood was drawn into EDTA containing tubes. After centrifugation plasma and cells were separated and frozen. DNA was prepared from the cells by standard procedures. Approved consent, according to the Helsinki Declaration, was given by the blood donors before samples were drawn. The samples from Amerindians were from two tribes, Kaingang and Guarani,39 the most numerous Indian tribes living in South Brazil. They are two different ethnic groups that differ culturally, and to some extent genetically, from each other. The language spoken by Kaingáng belongs to the Jê family, while the Guaraní's language belongs to the Tupí-Guaraní family. There are no Mestiços among the individuals presented here. The project was approved by the National Indian Council (FUNAI) and CONEP. The samples from Greenland were from individuals in West Greenland29 and East Greenland.40

MASP-2 concentration

The concentration of MASP-2 was measured as described in detail previously.23 In brief, the assay is based on microtitre wells coated with monoclonal anti-MASP-2 antibody (8B5). This antibody reacts with an epitobe in the serine protease domain of MASP-2, and does thus not react with MAp19. Samples diluted in buffer releasing MASP-2 from complexes with MBL or ficolins (1 M NaCl, 10 mM EDTA) are incubated in the wells, and a biotin-labeled monoclonal anti-MASP-2 (biotin-6G12) is subsequently used as detecting antibody, followed by europium-labeled streptavidin and reading by time-resolved fluorometry. Concentrations are read on standard curves made from an in-house standard plasma (calibrated against recombinant MASP-2) and each test includes three internal controls. The inter-assay coefficient of variation (%CV) was 8, 9 and 10% for the high (517 ng MASP-2/ml), medium (319 ng/ml) and low (170 ng/ml) internal controls, respectively.

MBL pathway activity

The ability of sera to activate complement factor C4 via the MBL pathway was tested as described previously.41 In brief, diluted serum samples are incubated in mannan-coated microtitre wells to allow for the binding of MBL-MASP complexes. This is performed in a buffer allowing for the binding of MBL without activating the MASPs. Subsequently, purified C4 is incubated in the wells at 37°C, allowing for activation of C4 and the covalent binding of C4b to the mannan surface. The amount of bound C4b is then detected with anti-C4 antibodies. This assay is strongly dependent on the concentration of MBL in the sample. Therefore, when evaluating C4b deposition capacity in sera with low MBL levels, recombinant MBL was added corresponding to a serum concentration of 5 μg/ml before testing the serum.

Sequencing of MASP2 exons

Genomic DNA was extracted and a PCR strategy was used to amplify the 12 exons (a to l) of MASP2. Each of these fragment were sequenced using a primer walking strategy to produce a full double stranded sequence. A detailed description of the gene with the boundaries of the exons is given in Stover et al.22 (in the following description the start of the polyadenylation signals used for the transcription of MAp19 and MASP-2 mRNA, are the 3′ ends of exon e and exon l, respectively). The following areas of the MASP2 gene were amplified:791 bases covering exons a to c (beginning 41 bases before exon a and ending 41 bases after the end of exon c); 414 bases covering exon d (beginning 55 bases before exon d and ending 226 bases after the end of exon d); 605 bases covering exon e (beginning 265 bases before exon e and ending 198 bases after the end of exon e); 1131 bp covering exon f and g (beginning 252 bases before exon f and ending 218 bases after the end of exon g); 400 bases covering exon h (beginning 133 bases before exon h and ending 197 bases after end of exon h); 457 bases covering exon i (beginning 286 bases before exon i and ending 91 bases after the end of exon i); 990 bases covering exon j and k (beginning 170 bases before exon j and ending 102 bases after the end of exon k); 1502 bases covering exon l (beginning 217 bases before exon l and ending 321 bases after the end of exon l). The sequencing reactions were performed at Lark Technologies Ltd., Essex, UK.

MASP2 genotyping

Genomic DNA was extracted from peripheral blood cells using the QIAamp DNA blood mini kit (Qiagen, Hilden, Germany).

A real-time PCR technique based on the 5′nuclease assay (TaqMan, Applied Biosystems, Foster City, CA, USA) in combination with minor-groove-binder (MGB) probes was used for screening the MASP2 gene for the five SNPs, p.R99Q, p.R118C, p.D120G, p.P126L and p.V377A and the 12 nucleotide duplication causing the amino acid duplication p.156_159dupCHNH (Table 1). We developed new TaqMan assays using primers amplifying parts of the MASP2 gene in combination with a set of MGB probes labeled with 6-carboxy-fluorescein (FAM) or VIC (proprietary dye from Applied Biosystems). Primer and probe sequences are shown in Table 3.

Table 3 Primer and probe sequences used for genotyping MASP2 gene

Cycling and conditions are identical for the analysis of all six polymorphisms of the MASP2 gene. PCR reactions containing 50 ng DNA, 900 nM primers, 200 nM probes and TaqMan Universal PCR Master Mix (Applied Biosystems) in a final volume of 25 μl were performed on a real-time PCR instrument (ABI Prism 7000). The PCR profile was 2 min at 50°C, 10 min at 95°C, followed by 40 cycles of 15 s at 92°C and 1 min at an annealing temperature of 60°C. End point reading of the fluorescence generated during PCR amplification was carried out on the ABI Prism 7000. Genotype assignments were obtained with the Sequence Detection System (SDS) software (Applied Biosystems). Owing to lack of DNA not all of the SNPs were tested in the Hong Kong samples.


Descriptive statistics for MASP-2 are presented by the median and minimum and maximum levels or interquartile range (IQR). Tests for location were performed using the Wilcoxon rank sum test or the Kruskal–Wallis test if more than two groups were compared. Measures of association were calculated using Spearman's rank correlation. P-values of less than 5% were considered significant.



mannan-binding lectin


single nucleotide polymorphism


MBL-associated serine protease-2








  1. 1

    Hargreaves DC, Medzhitov R . Innate sensors of microbial infection. J Clin Immunol 2005; 25: 503–510.

    CAS  Article  Google Scholar 

  2. 2

    Staros EB . Innate immunitynew approaches to understanding its clinical significance. Am J Clin Pathol 2005; 123: 305–312.

    CAS  Article  Google Scholar 

  3. 3

    Hoffmann JA, Kafatos FC, Janeway CA, Ezekowitz RA . Phylogenetic perspectives in innate immunity. Science 1999; 284: 1313–1318.

    CAS  Article  Google Scholar 

  4. 4

    Holmskov U, Thiel S, Jensenius JC . Collectins and ficolins: humoral lectins of the innate immune defense. Annu Rev Immunol 2003; 21: 547–578.

    CAS  Article  Google Scholar 

  5. 5

    Matsushita M, Fujita T . Ficolins and the lectin complement pathway. Immunol Rev 2001; 180: 78–85.

    CAS  Article  Google Scholar 

  6. 6

    Weis WI, Drickamer K, Hendrickson WA . Structure of a C-type mannose-binding protein complexed with an oligosaccharide. Nature 1992; 360: 127–134.

    CAS  Article  Google Scholar 

  7. 7

    Matsushita M, Endo Y, Taira S, Sato Y, Fujita T, Ichikawa N et al. A novel human serum lectin with collagen- and fibrinogenlike domains that functions as an opsonin. J Biol Chem 1996; 271: 2448–2454.

    CAS  Article  Google Scholar 

  8. 8

    Le Y, Lee SH, Kon OL, Lu J . Human L-ficolin: plasma levels, sugar specificity, and assignment of its lectin activity to the fibrinogen-like (FBG) domain. FEBS Lett 1998; 425: 367–370.

    CAS  Article  Google Scholar 

  9. 9

    Krarup A, Thiel S, Hansen A, Fujita T, Jensenius JC . L-ficolin is a pattern recognition molecule specific for acetyl groups. J Biol Chem 2004; 279: 47513–47519.

    CAS  Article  Google Scholar 

  10. 10

    Matsushita M, Fujita T . Activation of the classical complement pathway by mannose-binding protein in association with a novel C1s-like serine protease. J Exp Med 1992; 176: 1497–1502.

    CAS  Article  Google Scholar 

  11. 11

    Thiel S, Vorup-Jensen T, Stover CM, Schwaeble W, Laursen SB, Poulsen K et al. A second serine protease associated with mannan-binding lectin that activates complement. Nature 1997; 386: 506–510.

    CAS  Article  Google Scholar 

  12. 12

    Dahl MR, Thiel S, Matsushita M, Fujita T, Willis AW, Christensen T et al. MASP-3 and its association with distinct complexes of the mannan-binding lectin complement activation pathway. Immunity 2001; 15: 127–135.

    CAS  Article  Google Scholar 

  13. 13

    Sorensen R, Thiel S, Jensenius JC . Mannan-binding-lectin-associated serine proteases, characteristics and disease associations. Springer Semin Immunopathol 2005; 27: 299–319.

    CAS  Article  Google Scholar 

  14. 14

    Walport MJ . Complement First of two parts. N Engl J Med 2001; 344: 1058–1066.

    CAS  Article  Google Scholar 

  15. 15

    Walport MJ . Complement second of two parts. N Engl J Med 2001; 344: 1140–1144.

    CAS  Article  Google Scholar 

  16. 16

    Garred P, Larsen F, Seyfarth J, Fujita T, Madsen HO . Mannose-binding lectin and its genetic variants. Genes Immun 2006; 7: 85–94.

    CAS  Article  Google Scholar 

  17. 17

    Super M, Thiel S, Lu J, Levinsky RJ, Turner MW . Association of low levels of mannan-binding protein with a common defect of opsonisation. Lancet 1989; 2: 1236–1239.

    CAS  Article  Google Scholar 

  18. 18

    Worthley DL, Bardy PG, Mullighan CG . Mannose-binding lectin: biology and clinical implications. Intern Med J 2005; 35: 548–555.

    CAS  Article  Google Scholar 

  19. 19

    Thiel S, Frederiksen PD, Jensenius JC . Clinical manifestations of mannan-binding lectin deficiency. Mol Immunol 2006; 43: 86–96.

    CAS  Article  Google Scholar 

  20. 20

    Stengaard-Pedersen K, Thiel S, Gadjeva M, Moller-Kristensen M, Sorensen R, Jensen LT et al. Inherited deficiency of mannan-binding lectin-associated serine protease-2. N Engl J Med 2003; 349: 554–560.

    CAS  Article  Google Scholar 

  21. 21

    Lozano F, Suarez B, Munoz A, Jensenius JC, Mensa J, Vives J et al. Novel MASP2 variants detected among North African and Sub-Saharan individuals. Tissue Antigens 2005; 66: 131–135.

    CAS  Article  Google Scholar 

  22. 22

    Stover C, Endo Y, Takahashi M, Lynch NJ, Constantinescu C, Vorup-Jensen T et al. The human gene for mannan-binding lectin-associated serine protease-2 (MASP-2), the effector component of the lectin route of complement activation, is part of a tightly linked gene cluster on chromosome 1p362-3. Genes Immun 2001; 2: 119–127.

    CAS  Article  Google Scholar 

  23. 23

    Moller-Kristensen M, Jensenius JC, Jensen L, Thielens N, Rossi V, Arlaud G et al. Levels of mannan-binding lectin-associated serine protease-2 in healthy individuals. J Immunol Methods 2003; 282: 159–167.

    CAS  Article  Google Scholar 

  24. 24

    Carlsson M, Sjoholm AG, Eriksson L, Thiel S, Jensenius JC, Segelmark M et al. Deficiency of the mannan-binding lectin pathway of complement and poor outcome in cystic fibrosis: bacterial colonization may be decisive for a relationship. Clin Exp Immunol 2005; 139: 306–313.

    CAS  Article  Google Scholar 

  25. 25

    Stover C, Barrett S, Lynch NJ, Barker JN, Burden D, Trembath R et al. Functional MASP2 single nucleotide polymorphism plays no role in psoriasis. Br J Dermatol 2005; 152: 1313–1315.

    CAS  Article  Google Scholar 

  26. 26

    Cedzynski M, Szemraj J, Swierzko AS, Bak-Romaniszyn L, Banasik M, Zeman K et al. Mannan-binding lectin insufficiency in children with recurrent infections of the respiratory system. Clin Exp Immunol 2004; 136: 304–311.

    CAS  Article  Google Scholar 

  27. 27

    Olesen HV, Jensenius JC, Steffensen R, Thiel S, Schiøtz PO . The mannan-binding lectin pathway and lung disease in cystic fibrosis – dysfunction of mannan-binding-lectin-associated serine protease 2 (MASP-2) may be a major modifier. Clin Immunol 2006; 121: 324–331.

    CAS  Article  Google Scholar 

  28. 28

    García-Laorden MI, García-Saavedra A, Rodríguez de Castro F, Violán JS, Rajas O, Blanquer J et al. Low clinical penetrance of mannose-binding lectin–associated serine protease 2 deficiency. J Allergy Clin Immunol 2006; 118: 1383–1386.

    Article  Google Scholar 

  29. 29

    Grunnet N, Steffensen R, Jersild C . Increased frequency of HLA-DRw14b(w6)-associated RFLP in Greenlanders of Eskimo origin. Tissue Antigen 1991; 373: 127–129.

    Article  Google Scholar 

  30. 30

    Helgason A, Palsson G, Pedersen HS, Angulalik E, Gunnarsdottir ED, Yngvadottir B et al. mtDNA variation in Inuit populations of Greenland and Canada: migration history and population structure. Am J Phys Anthropol 2006; 130: 123–134.

    Article  Google Scholar 

  31. 31

    Ip WK, Chan KH, Law HK, Tso GH, Kong EK, Wong WH et al. Mannose-binding lectin in severe acute respiratory syndrome coronavirus infection. J Infect Dis 2005; 191: 1697–1704.

    CAS  Article  Google Scholar 

  32. 32

    Garred P, Madsen HO, Marquart H, Hansen TM, Sorensen SF, Petersen J et al. Two edged role of mannose binding lectin in rheumatoid arthritis: a cross sectional study. J Rheumatol 2000; 27: 26–34.

    CAS  Google Scholar 

  33. 33

    Aggarwal BB . Signalling pathways of the TNF superfamily: a double-edged sword. Nat Rev Immunol 2003; 3: 745–756.

    CAS  Article  Google Scholar 

  34. 34

    Wallis R, Dodd RB . Interaction of mannose-binding protein with associated serine proteases: effects of naturally occurring mutations. J Biol Chem 2000; 275: 30962–30969.

    CAS  Article  Google Scholar 

  35. 35

    Thielens NM, Cseh S, Thiel S, Vorup-Jensen T, Rossi V, Jensenius JC et al. Interaction properties of human mannan-binding lectin (MBL)-associated serine proteases-1 and -2, MBL-associated protein 19, and MBL. J Immunol 2001; 166: 5068–5077.

    CAS  Article  Google Scholar 

  36. 36

    Stover CM, Thiel S, Thelen M, Lynch NJ, Vorup-Jensen T, Jensenius JC et al. Two constituents of the initiation complex of the mannan-binding lectin activation pathway of complement are encoded by a single structural gene. J Immunol 1999; 162: 3481–3490.

    CAS  Google Scholar 

  37. 37

    Gregory LA, Thielens NM, Matsushita M, Sorensen R, Arlaud GJ, Fontecilla-Camps JC et al. The X-ray structure of human mannan-binding lectin-associated protein 19 (MAp19) and its interaction site with mannan-binding lectin and L-ficolin. J Biol Chem 2004; 279: 29391–29397.

    CAS  Article  Google Scholar 

  38. 38

    Gal P, Harmat V, Kocsis A, Bian T, Barna L, Ambrus G et al. A true autoactivating enzyme Structural insight into mannose-binding lectin-associated serine protease-2 activations. J Biol Chem 2005; 280: 33435–33444.

    CAS  Article  Google Scholar 

  39. 39

    Weg-Remers S, Brenden M, Schwarz E, Witzel K, Schneider PM, Guerra LK et al. Major histocompatibility complex (MHC) class III genetics in two Amerindian tribes from southern Brazil: the Kaingang and the Guarani. Hum Genet 1997; 100: 548–556.

    CAS  Article  Google Scholar 

  40. 40

    Eiberg H, Nielsen IM . Linkage studies of Cholestasis familiaris Groenlandica/Byler-like disease, with polymorphic protein and blood group markers. Human Heredity 1993; 43: 250–256.

    CAS  Article  Google Scholar 

  41. 41

    Thiel S, Møller-Kristensen M, Jensen L, Jensenius JC . Assays for the functional activity of the mannan-binding lectin pathway of complement activation. Immunobiology 2002; 205: 446–454.

    CAS  Article  Google Scholar 

Download references


We thank Dr Christine Gaboriaud, Grenoble, France for discussions on the possible structural influences of the polymorphisms. We are grateful for the excellent technical assistance from Hanne Nielsen, Annette Hansen, Louise Jakobsen and Lisbeth Jensen. This study was supported by Danish Research Council and Novo Nordic Foundation.

Author information



Corresponding author

Correspondence to S Thiel.

Additional information

Financial interests

ST and JCJ have financial interest in NatImmune A/S, a biotech company exploring the possibilities of therapy with proteins of the innate immune system.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Thiel, S., Steffensen, R., Christensen, I. et al. Deficiency of mannan-binding lectin associated serine protease-2 due to missense polymorphisms. Genes Immun 8, 154–163 (2007).

Download citation


  • humans
  • complement system
  • single nucleotide polymorphisms
  • immune deficiency

Further reading