Analysis of var Gene Transcription Pattern Using DBLα Tags

The Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1) antigens, which are encoded by a multigene family called var genes, are exported and inserted onto the surface of the infected erythrocytes. PfEMP1 plays a key role in the pathogenesis of severe malaria and are major targets of naturally acquired immunity. Studying the expression pattern of var genes in P. falciparum clinical isolates is crucial for understanding disease mechanism and immunity to malaria. However, var genes are highly variable, which makes it difficult to study their expression in clinical isolates obtained directly from malaria patients. In this chapter, we describe an approach for analysis of var gene expression that targets a region referred to as DBLα tag, which is relatively conserved in all var genes.


Introduction
var is a multigene family that encodes Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1). There are about 60 var genes in the haploid genome of each isolate [1] and there is minimal repertoire conservation between genomes. Generally, var genes are made up of two exons. Exon 1 is highly variable and encodes for the part of the PfEMP1 that is exposed on the infected erythrocyte surface. This part of PfEMP1 is composed of a combination of multiple domains of Duffy binding-like (DBL) and cysteine-rich interspersed region (CIDR) domains and the N-terminal segment (NTS) [1,2]. Even though some sequence homology can be identified in the DBL domains, these homology blocks are flanked by hypervariable regions that contain few conserved residues and no particular structural features [3,4]. Exon 2 on the other hand is relatively conserved and is made up of the acidic terminal segment (ATS).
Serological work has also supported the importance of PfEMP1 in natural infections [13][14][15][16][17]. Clinical studies present the challenge that the infecting isolates do not have their genomes sequenced. This, together with the diversity of var genes, makes it difficult to study var gene transcription in clinical isolates. However, a number of studies have shown that var genes have a semiconserved head structure [1,2]. Gardner et al. showed that in the 3D7 genome, the DBL alpha (DBLα) domain occurred in most of the var genes and formed part of the semiconserved head structure [1]. Taking advantage of this, Taylor et al. [18] designed degenerate primers that can be used in amplifying a small region within the DBLα domain referred to as the DBLα tag (Fig. 1). Studies have demonstrated that the DBLα tag sequence can provide functional information related to the full-length var sequence [19].
var genes containing DBLα-tag sequence with a reduced number of cysteine residues have been shown to predominantly fall under group A and to be preferentially transcribed by isolates from children with severe malaria and low host immunity [7,8,[20][21][22]. Here, we describe an approach that we have used to determine var gene transcription using the DBLα tag. We describe the use of DBLα sequencing from clinical isolates and counting of the var gene sub-groups as a proportion of the var, as well as the use of var expression homogeneity

Sample Preparation
To prepare infected erythrocyte pellet for RNA extraction from clinical samples:

2.
Transfer the blood to a sterile 15 mL centrifuge tube under a laminar flow hood.

3.
To separate plasma from the cellular components, centrifuge the blood at 440 × g for 5 min and remove supernatant (plasma).

4.
Resuspend the remaining cells in 5 mL buffered incomplete RPMI 1640 medium.

5.
Carefully layer resuspended cells on 3 mL Lymphoprep in 15 mL centrifuge tube and centrifuge at 440 × g for 20 min to separate PBMCs (see Note 1 ).

6.
After centrifugation, remove the distinct PBMC layer at the interface of the medium and Lymphoprep using a wide mouth Pasteur pipette (Fig. 2).

7.
Wash remaining cells in 10 mL "yellow" RPMI by centrifugation at 440 × g for 5 min.

8.
To remove granulocytes from the remaining erythrocytes, make a 40% erythrocyte suspension by adding 1.5 times of warm "yellow" RPMI to the cell pellet obtained in step 7 and an equal amount of warm Plasmion.

9.
Mix thoroughly and let the tube stand in a water bath at 37 °C for 10 min.

10.
Collect supernatant containing granulocytes into a separate tube and wash remaining erythrocytes using 10 mL of warm "yellow" RPMI by centrifuging at 440 × g for 5 min.

1.
Transfer the infected erythrocyte pellet from Subheading 3.1 into a 15 mL centrifuge tube.
1 Lymphoprep or any other white blood cell depletion reagent can be used in order to obtain pure erythrocytes.
Andisi and Abdi Page 4

cDNA Synthesis
Two microliters of extracted RNA is used to make (cDNA) using the Superscript III kit according to the manufacturer's protocol.

1.
To remove any contaminating DNA, digest RNA using 1 μL of Ambion DNAse enzyme for 20 min at 37 °C.

2.
To remove the DNase, add 3 μL of Ambion DNase inactivation reagent to the reaction, mix and stand for 2 min at room temperature.

3.
Centrifuge at 9408 × g for 1.5 min and transfer two aliquots of 8 μL of supernatant containing RNA into clean PCR strip tubes.

4.
Next, reverse transcribe all the RNA to make the first strand using the SuperScript III kit in the presence of random hexamers and dNTPs according to the manufacturer's instructions (see Note 3 ).

DBLα Tag Amplification
To capture the majority of the var genes being expressed by each P. falciparum isolate, degenerate primers DBLαAF (GCACGMAGTTTYGC) and DBLαBR (GCCCATTCSTCGAACCA) targeting the semiconserved DBLα-tag sequence are used for amplification [18].

1.
Prepare a 25 μL PCR reaction mixture including a 2 μL template from the cDNA reaction.

2.
Run the PCR reaction using the following conditions: denaturation 95 °C, annealing 42 °C, extension 65 °C, and a final extension of 65 °C for 30 cycles on thermocycler.

3.
Prepare a 2% agarose gel by mixing 2 g agarose powder with 100 mL 1 × TAE buffer in a microwavable flask and microwave for 30 s or until the powder melts completely.

4.
After cooling to about 50 °C, add a preferred gel stain (see Note 4 ) to a final concentration of approximately 0.2-0.5 μg/mL and pour into a gel tray with the well comb in place.

5.
Using a suitable loading dye, load and fractionate 5 μL of the amplified PCR product on the agarose gel for 90 min at 90 kV.

6.
View the stained gel under ultraviolet light in a transilluminator for the expected product size of 350-450 bp.

1.
To prepare the column, place empty microspin columns in eppendorf tubes/ collection tubes.

DBLα Sequence Assembly, Classification, and Counting
We use two main approaches to classify the DBLα tags. These approaches and algorithms were developed and published by Bull et al. [23,24]. In the first approach, DBLα tags are classified using distinct sequence features (Fig. 3) into six groups based on the number of cysteine amino acid and the presence of semiconserved motifs REY/MFK. These motifs occur in a mutually exclusive manner among the short DBLα sequences containing two cysteines, at the positions of limited variability (PoLV) [24]. This is referred to as the Cys/ PoLV or CP grouping.

1.
Following base-calling using Phred software, remove/clip low quality ends and assemble the forward and reverse reads.

2.
Translate to obtain an open reading frame and capture DBLα tags by use of semiconserved features including DIGDI, VW, WW, and PQYLR motifs as described in Bull et al. [24].

3.
Exclude any sequences that encode peptides shorter than 100 amino acids (i.e., ≤300 bp) and remove the constitutively expressed var1 sequences from the analysis.

4.
Classify the tags obtained into Cys2 for those containing two cysteines, Cys4 for those with four and CysX for those containing one, three, five, or six cysteines.

6.
An alternative approach is the use of a network of recombining sequences to define blocks of sequences that tend to recombine with each other. This algorithm uses block sharing groups (BS groups) made up of polymorphic sequence blocks together with the number of cysteines in the DBLα tag [23].

7.
Sequences that fall into block-sharing Group 1 and have two cysteines (BS1_Cys2) belong to group A-like var genes (see Note 6 ).

8.
Count all the reads per P. falciparum isolate falling into each of these groups and express as a proportion of the total number of reads obtained for the isolate.  [7,20]. However, deeper sequencing provides better estimates of diversity in expression. 6 Bs1_Cys2 sequences are defined as the "group A-like" var genes because they tend to be group A vars. However, a different group which captures bs1_Cys2 sequences, in addition to Cys/PoLV group 1 can be derived. We refer to these as bs1_cys2_cp1. All sequences falling in the Cys/PoLV group 1 are known to also fall under the group A var genes but not all bs1_Cys2_cp1are group A var genes. Additional groups can be defined, including sequences that fall into the BS group2 and Cys/PoLV group2. These are referred to as BS2_CP2.
Andisi and Abdi Page 8 Methods Mol Biol. Author manuscript; available in PMC 2022 September 08.

9.
Following DBLα classification, var expression homogeneity (VEH) index can also be calculated. VEH is defined as the extent to which a small number of var gene sequences dominate an isolates expression profile [6]. VEH is calculated using the Simpsons diversity index defined here as the sum of squares of the frequencies of each sequence type in the var profile. Thus, the lower VEH the more heterogeneous an isolate's var expression profile. Depletion of white blood cells. Illustration of depletion of white blood cells from whole blood using Lymphoprep density gradient medium. Figure demonstrates layering of blood before centrifugation and the distinct mononuclear cells, RPMI/plasma and erythrocyte layers after centrifugation Andisi   DBLα tag sequence features. Location of sequence features used in classification of DBLα tags demonstrated using five DBL α sequences derived from clinical P. falciparum isolates.
The anchor points are in blue, Positions of Limited Variability (PoLV) are in grey and number of cysteines in green [24] Andisi and Abdi Page 13 Methods Mol Biol. Author manuscript; available in PMC 2022 September 08.