Copy number analysis of complement C4A, C4B and C4A silencing mutation by real-time quantitative polymerase chain reaction

Abstract

Low protein levels and copy number variation (CNV) of the fourth component of human complement (C4A and C4B) have been associated with various diseases. High-throughput methods for analysing C4 CNV are available, but they commonly do not detect the most common C4A mutation, a silencing CT insertion (CTins) leading to low protein levels. We developed a SYBR® Green labelled real-time quantitative polymerase chain reaction (qPCR) with a novel concentration range approach to address C4 CNV and deficiencies due to CTins. This method was validated in three sample sets and applied to over 1600 patient samples. CTins caused C4A deficiency in more than 70% (76/105) of the carriers. Twenty per cent (76/381) of patients with a C4A deficiency would have been erroneously recorded as having none, if the CTins had not been assessed. C4A deficiency was more common in patients than a healthy reference population, (OR = 1.60, 95%CI = 1.02-2.52, p = 0.039). The number of functional C4 genes can be straightforwardly analyzed by real-time qPCR, also with SYBR® Green labelling. Determination of CTins increases the frequency of C4A deficiency and thus helps to elucidate the genotypic versus phenotypic disease associations.

Figure 1. Method validation with selected samples.

A. Immunophenotyping. The gel is skewed in the middle, leading to lower position of protein band levels on the left. All samples were analysed in a replicate run. B. Real-time quantitative PCR (qPCR) for copy number variation. The y-axis depicts the linear view of the fluorescence rate (from 0 to 0.6 in the full picture, from 0 to 0.06 in the magnification) and the x-axis the number of cycles (from 0 to 30 in the full picture and 15 to 21 in the magnification). Each curve represents the mean of two replicates of a sample. The lowest horizontal line represents non-template controls (sterile water), negative control and samples with zero copies of C4B having zero fluorescence due to undetectable amounts of DNA (TX-2144 and TX-2147). The curves from left to right depict samples with C4B CNV 3, 2 and 1 (TX-1586, TX-2170 and TX-2158, TX-2209 and TX-2284, respectively). The number of cycles at which the fluorescence curve cuts the threshold (the red horizontal line) is recorded; the greater amount of genes indicates the lower number of cycles surpassing the threshold. C4A and CTins qPCR runs resulted in similar output. C. Parallel results of C4 analyses. Functional C4A CNV was assessed by reducing the amount of CTins from C4A CNV.

Figure 2. Frequencies of C4 deficiencies in different populations.

The frequencies (%) of phenotypic C4A deficiency (functional C4 copy number <2, CTins reduced from C4A), C4A deficiency (copy number <2) and C4B deficiency (copy number <2) are shown. The populations are Finnish patients from the current study with unambiguous C4 qPCR results (n = 1618), Finnish (n = 149) , Hungarian (n = 118) , U.K. (n = 719) , Spanish (n = 460) and Dutch (n = 104) general population samples.