Analysis of Tandem Repeat Patterns in Nlrc4 using a Motif Model

Published Online December 2012 in MECS DOI: 10.5815/ijitcs.2013.01.06

The advancement of computational tools such as databases [1-5], modelling [6-8], decision support systems [9], simulation [10-12], visualization techniques [13-16], and web servers [17-20] have revolutionized the field of bioinformatics and molecular biology with the breakthrough in data processing and analysis. These computational tools received wide application in various specialized disciplines in molecular biology, including systems biology [21-22], genome analysis [23-24], population genetics [25], structural bioinformatics [26-27], and phylogenetics [28-29]. Various biological data have been processed efficiently and accurately with the powerful computational tools at hand. Bioinformaticians and computer scientists have employed these tools to study protein functions [30-34], gene expression [35-38], regulatory gene identification [39-41], and the evolution of gene [42-43].

This research adopts a computational analysis to identify the tandem repeat patterns in Nlrc4 gene. Nlrc4 gene encodes for NLRC4 protein, which is a caspase recruitment domain (CARD) containing, NODlike receptor (NLR) family member that plays a crucial role in innate sensing of pathogens. It is one of the known genes that involves in the formation of inflammasome, which is a molecular complex that mediates the cytosolic pathogen detection and signal cascading [44,46-47]. Notably, the activation of NLRC4 inflammasome by Salmonella and Legionella requires flagellin and functional type III or IV secretion systems [45,49]. Other route of activation is possible as S. flexneri, which is a microorganism lacking of flagellin, can be sensed by NLRC4 [51]. An NLRC4 inflammasome is self-oligomerized and formed by domain interactions between its CARD domain with the CARD domain of pro-caspase-1 [48]. This interaction leads to the cleavage of pro-caspase-1 into a form of activated caspase-1, which causes rapid cell death [50]. Cell death caused by caspase-1 activation will stimulate immune response to eliminate the dead cells.

This research aims to analyze the tandem repeat patterns in Nlrc4 gene which encodes for NLRC4 inflammasome. Tandem repeats are unstable genomic elements which are repeated several times. For instance, the nucleotide sequence CGCGCGCG is a tandem repeat of nucleotides cytosine (C) and guanine (G) for four times; and sequence AGTAGTAGT is a tandem repeat of adenine (A), guanine (G) and thymine (T) for three times. Tandem repeats are ubiquitous in the genome of organisms across species, and the mutations in the regions of tandem repeat have phenotypic consequences [52], such as diseases. In inflammasome aggregates, tandem repeats can be found in the leucine-rich repeat (LRR) region. LRR is also responsible for other non-immunity role, such as the development and functions of neural circuits [53]. The emergence of tandem repeat drives the evolution of organisms, and multiple copies of gene in the form of tandem repeat allow cells to survive under growth-limiting milieu [54]. Hence, analysis of tandem repeats may provide insights in the pathogenesis and the evolution of an organism.

This paper elaborates the methods used in the Section 2. A motif model is provided in this section. Results of tandem repeat patterns are provided in Section 3, with the analysis of the distribution patterns and the signaling pathways. Finally, Section 4 concludes the findings of the tandem repeat patterns of Nlrc4 gene.

• II.    Methods

Sequence data of Nlrc4 gene of human was retrieved from NCBI GenBank. The UCSC genome browser [55]

was used to visualize the locus of Nlrc4 gene on the chromosome. KEGG database [56] was used to extract and match the regulatory and metabolic pathways of NLR signaling pathway. We used mreps algorithm [57] to compute the statistically expected repeats. A motif

M model was used to identify the motif 0 from the nonmotif 0 [58]. The non-motif is the background motif which does not display the pattern of tandem repeat.

The motif model [58] is depicted in (1) below:

	"oWM...	0M " _, w
	B MM “ a 1,o “ a 1,1 “ a 1,2	M • • “a 1, w
0 = { 0 B , 0 M } =	B MM “ a 2,0 “ a 2,1 “ a 2,2	M . . “ a 2, w
	e B ,o “ 0M, 1 “ M 2 ■	'" M • ' “aj , w	(1)

The subsequence of length W at position j in a sequence i is denoted by S^ij . Let a be the symbol that

MB occurs at a position k of either 0 or 0  ; let the position k be 1 — k — W, and L be a non-empty set for the length of nucleotide sequence. The conditional probabilities that Sij is found using the motif model are [58]:

WL

P M ( S^ij ) =

П П “ M ) I ⁽ S j^{k - 1} = ^a )

k = 1     a = 1

the conditional probabilities that S^ij is found using the non-motif model are computed as [58]:

Let ^X be the prior probability of motif occurrence in the gene sequences of Nlrc4 gene. The motif occurrence probability Z at position j in sequence i is derived from the formula below [58]:

WmS"

Z. = ( X P „ S ⁱ ) + (1 - X ) P b S“          ₍₄₎

Furthermore, a genuine motif will fulfill the following condition [58]:

log ( P „( S »)/ P b ( S ^j ) ) > log[(1 - X ) X ]   ₍₅₎

In gene analysis, we need to take the phenomenon of bogus identification of gene motif into account. Such bogus motif shall be discarded. A particular count of motif is taken to be a pseudo-count by [58]:

L pscak = ^ “b,k +1 Pa / b b=1                                            (6)

Where P_a/b is the BLOSUM substitution probability for amino acid a from the observed amino acid b . Pseudo-count of motif will be discarded in the finalized count.

• III.    Results and Discussion

  Nlrc4 is a gene with a length of 3385 base pairs (bp), with the nucleotide composition of A=28.98%, T=25.91%, G=23.40%, and C=21.71%. The gene locus is depicted in Fig. 1.

WL

П П <) I ⁽ '-^{k - 1} = ^a )

P b ( S i ) = ^k = ¹    a = ¹                                 (3)

Fig. 1: Nlrc4 on the chromosome

Fig. 1 shows that Nlrc4 is located at chromosome 2. The lower panel of the figure shows that there are extensive repeating elements in the gene region. We identified a particular sequence to be exhibiting a pattern of tandem repeats when a stretch of mononucleotides repeat at least 4 times, and twice for di-, tri-, tetra-, penta-nucleotides. Table 1 illustrates the occurrence and relative frequency of tandem repeat motifs. The relative frequency was computed as per 1k bp.

Table 1: The occurrence and relative frequency of tandem repeat motifs

Tandem Repeat Motif	Occurrence	Relative Frequncy
Mononucleotide	53	179.41
Dinucleotide	111	375.74
Trinucleotide	33	111.71
Tetranucleotide	0	0
Pentanucleotide	0	0

The results in Table 1 show that the dinucleotide tandem repeat is the most prevalent repeat pattern in Nlrc4 gene. Its relative frequency is double than that of the mononucleotide tandem repeat. We observed that tandem repeats have declined as the order of nucleotide number increases. There are 33 occurrences (relative frequency=111.71) of trinucleotide tandem repeats, whereas none was found in tetranucleotide and pentanucleotide. The absence of tetranucleotide and pentanucleotide tandem repeats can be explained by the small size of Nlrc4 gene.

The distribution of tandem repeats was recorded for each 1kbp of Nlrc4 gene, as shown in Table 2. The numbers in each column indicate the number of occurrence.

Table 2: The distribution of tandem repeats across the Nlrc4 gene region

Tandem Repeat Motif	1-1kbp	1kbp-2kbp	2kbp-3385bp
Mononucleotide	19	9	25
Dinucleotide	34	39	38
Trinucleotide	10	8	15

From Table 2, it was observed that both mononucleotide and trinucleotide tandem repeat exhibit a pattern of reducing occurrence of tandem repeats in the mid-locus region of Nlrc4 gene. However, the reducing number of trinucleotide tandem repeats in the mid-locus region is not statistically significant as the difference from the other regions is not large. As for dinucleotide tandem repeat distribution, we observed an even distribution in all regions of Nlrc4 gene. Besides, we also identified the number of tandem repeat occurrence at the vicinity of 5’ and 3’ region of Nlrc4 gene. The vicinity is defined as 100 bp downstream and upstream of 5’ and 3’ region, respectively. The observed occurrence of tandem repeat was low in 5’ and

3’ vicinity of Nlrc4 gene. Hence, tandem repeats may not have significant impacts on the promoter binding and pathogen sensing in a case where they are mutated in these regions. The results are shown in Table 3.

Table 3: The occurrence of tandem repeat at the vicinity of 5’ and 3’ region of Nlrc4 gene

Tandem repeat motif	5’ vicinity	3’ vicinity
Mononucleotide	3	2
Dinucleotide	3	3
Trinucleotide	0	0

We observed a few tandem repeats which occur in an order higher than 4, as given in Table 4.

Table 4: The tandem repeats occur in an order > 4

Repeat motif	Start locus	End locus
(A) 6	178	183
(A) 13	3373	3385
(A) 5	453	457
(T) 6	3063	3068
(T) 6	3180	3185
(A) 5	1392	1396
(A) 5	1887	1891
(A) 5	2637	2641
(A) 5	2857	2861
(A) 5	3018	3022
(A) 5	3084	3088
(T) 5	541	545
(T) 5	655	659
(T) 5	910	914
(G) 5	1166	1170
(G) 5	2127	2131

The relative frequency of the high order (>4) tandem repeats is only 54.16, implying that for every 1000bp, there are only about 54 bp that are the constituent nucleotide of a higher order tandem repeats. Notice also that all of the high order tandem repeats listed in Table 4 are mononucleotide tandem repeats. The percentage of high order mononucleotide tandem repeat is 30.20% of the total number of mononucleotide tandem repeat occurrence, which is considerably high (for mononucleotide only). However, high order tandem repeats are always scarce in general, as it was also reported by other researcher, such as the similar findings reported by Gur-Arie et al. [59] in their analysis on E. coli .

The NLR signaling pathways of Nlrc4 protein product were identified and matched using KEGG database [56]. These pathways are illustrated in Fig. 2.

Fig. 2: NLR signaling pathway of NLRC4

Fig. 2 demonstrates that the signaling pathways are originated from bacteria, antiviral chemical compounds, danger signals derived from host cells, and xenogenous compounds such as silica. NOD-like receptor (NLR) which can sense a wide variety of pathogen molecular patterns and danger molecular patterns has four major pathways, which lead to either cell death or the activation of caspase-1 to induce the expression of pro-inflammatory cytokines. The first pathway (the upper panel in Fig. 2) is a signaling pathway that detects the peptidoglycan of bacteria. It is an inflammasome-independent pathway mediated by NOD1 and NOD2 proteins, which culminates in the activation of transcription factor NF- κ B. NF- κ B plays an important role in multiple pathogen sensing mechanisms [60-63], leading to the expression of pro-inflammatory cytokines and chemokines. The second pathway (the middle panel in Fig. 2) demonstrates a cascade of pathways mediated by inflammasome which detects toxins produced by bacteria. This pathway is not totally independent from the first pathway because the inflammasome can be activated by bacterial cell wall component muramyl dipeptide (MDP). This pathway culminates in either cell death or the activation of pro-inflammatory cytokine, which leads to the release of IL-1β and IL-18 to extracellular space. The third pathway is mediated by reactive oxygen species (ROS) that present in cytosol of cells in the condition of cellular stress. The presence of ROS can activate inflammasome, which results in the cleavage of pro-IL-1β to a mature form of IL-1β [64]. The fourth, and the last, pathway (the lowest panel in Fig. 2) is mediated by bacterial flagellin that activates NLRC4 inflammasome. Similar to the second signaling pathway, the cellular outcome of this NLRC4 inflammasome-activated pathway culminates in either cell death or the release of active pro-inflammatory cytokines.

• IV.    Conclusion

Nlrc4 is a gene that encodes for NLRC4 protein (inflammasome) which is implicated in pathogen sensing and the corresponding immune responses. This research applied a computational approach to analyze the tandem repeat patterns in Nlrc4 gene. We found that the dinucleotide tandem repeat is the most prevalent repeat pattern in Nlrc4 gene. In our analysis, the occurrence of tandem repeat was low in 5’ and 3’ vicinity of Nlrc4 gene, implying that the mutated tandem repeats may not have significant impacts on the promoter binding and pathogen sensing. Although as much as 30.20% of mononucleotide tandem repeats occur for more than 4 times consecutively, the high order tandem repeats are always scarce. The signaling pathway demonstrates that the normal expression of Nlrc4 gene is essential for the activation of immune response in the presence of pathogens. Particular attention needs to be paid to the mono-, di-, and trinucleotide tandem repeat mutation in the diagnosis of Nlrc4-implicated diseases. This research provides an insight into the therapeutic strategies and drug designs for Nlrc4-related diseases.