Proteomic analysis of microparticles isolated from malaria positive blood samples.

BACKGROUND: Malaria continues to be a great public health concern due to the significant mortality and morbidity associated with the disease especially in developing countries. Microparticles (MPs), also called plasma membrane derived extracellular vesicles (PMEVs) are subcellular structures that are generated when they bud off the plasma membrane. They can be found in healthy individuals but the numbers tend to increase in pathological conditions including malaria. Although, various studies have been carried out on the protein content of specific cellular derived MPs, there seems to be paucity of information on the protein content of circulating MPs in malaria and their association with the various signs and symptoms of the disease. The aim of this study was therefore to carry out proteomic analyses of MPs isolated from malaria positive samples and compare them with proteins of MPs from malaria parasite culture supernatant and healthy controls in order to ascertain the role of MPs in malaria infection.
METHODS: Plasma samples were obtained from forty-three (43) malaria diagnosed patients (cases) and ten (10) healthy individuals (controls). Malaria parasite culture supernatant was obtained from our laboratory and MPs were isolated from them and confirmed using flow cytometry. 2D LC-MS was done to obtain their protein content. Resultant data were analyzed using SPSS Ver. 21.0 statistical software, Kruskal Wallis test and Spearman's correlation coefficient r.
RESULTS: In all, 1806 proteins were isolated from the samples. The MPs from malaria positive samples recorded 1729 proteins, those from culture supernatant were 333 while the control samples recorded 234 proteins. The mean number of proteins in MPs of malaria positive samples was significantly higher than that in the control samples. Significantly, higher quantities of haemoglobin subunits were seen in MPs from malaria samples and culture supernatant compared to control samples.
CONCLUSION: A great number of proteins were observed to be carried in the microparticles (MPs) from malaria samples and culture supernatant compared to controls. The greater loss of haemoglobin from erythrocytes via MPs from malaria patients could serve as the initiation and progression of anaemia in P.falciparum infection. Also while some proteins were upregulated in circulating MPs in malaria samples, others were down regulated.

Background

Microparticles (MPs) also called plasma membrane derived extracellular vesicles (PMEVs) are a heterogeneous group of small sub-membrane fragments or membrane coated vesicles shed from the plasma membrane of various cells during normal cellular activities like growth, senescence, proliferation and apoptosis [1]. MPs carry proteins, lipids and nucleic acids from host cells and are means of intercellular communication and it has been shown that analysis of MPs from blood samples can provide information about the state and progression of a particular disease or condition [2].

Malaria is caused by five species of the genus Plasmodium which is a unicellular protozoan parasite. The disease is a major cause of mortality and morbidity in many developing countries especially in Sub-Saharan Africa. It is estimated 3.4 billion people worldwide risk being infected with malaria in 104 countries [3]. Complications associated with malaria infection particularly in severe malaria include fever/chills, coagulopathy and anaemia among other symptoms. At the molecular level, up-regulation of certain cytokines is also thought to relate to malaria associated high fever [4].

The degree of anaemia experienced in malaria does not always correspond to the parasitaemia level [5]. This is partly caused by a mild bone marrow suppression of erythrocyte production and the collection of complement containing complexes on erythrocyte surfaces after infection which promotes splenic removal of these erythrocytes. Lysis of both infected and uninfected erythrocytes is also considered to be a contributing factor [5]. Severe malaria caused by Plasmodium falciparum is considered to be associated with the dysregulation of the coagulation system which include endothelial damage, lower levels of anticoagulation and the release of pro-coagulant MPs [6].

To this end, malaria has been associated with an increase in the level of circulating plasma MPs and plasma concentrations of endothelial MPs (EMPs) which may be proportional to disease severity [7, 8]. The role of infected erythrocyte-derived MPs in cellular communication has been investigated but the protein content (proteomic analysis) of MPs isolated in malaria is yet to be explored [9]. Proteomic analysis on circulating MPs obtained from plasma of malaria positive blood samples once explored will give a general idea of the protein and protein groups borne by these MPs that may influence the pathophysiology of malaria infection. This study therefore sought to examine the protein composition of plasma MPs of malaria samples and comparing them with proteins of MPs from healthy controls in order to explore their effect on the pathogenesis of malaria and the possible linkage of circulating plasma MPs to malaria anaemia.

Existing literature indicate that elevated MPs levels have been seen in cancer, sepsis, pulmonary hypertension, idiopathic thrombocytopenic purpura and atherosclerosis [10]. Researchers also contend that increased endothelial microparticle level correlating with disease severity has been seen in malaria [11]. Again studies in mice models indicate that microparticles contributed to induction of systemic inflammation [12, 13]. Others have shown that MPs released after malaria infection which are primarily erythrocyte-derived are capable of activating macrophage through toll-like receptors (TLR) and may enhance infectivity as their count elevates and investigations show they contain parasite components some of which promote pathogen invasion of erythrocytes [14].

Regev-Rudzki et al. (2013) postulated that MPs released in malaria are also capable of activating the blood–brain barrier which exacerbates inflammation [15]. The perplexing feature of malarial anemia which is increased clearance of uninfected erythrocytes can also be attributed to the release of parasite antigens in MPs during entry in erythrocytes. These erythrocyte-adhesive proteins probably adhere to erythrocytes resulting in IgG and complement binding which promotes their elimination from peripheral circulation [16]. Furthermore, Schorey et al. stated that MPs released from P. falciparum-infected cells are able to modulate hosts immune response thereby impairing surveillance [17].

Proteomic analysis on circulating plasma MPs obtained from plasma of malaria positive blood samples once explored will give a general idea of the protein and protein groups borne by these MPs thereby influencing the pathophysiology of malaria infection. This study seeks to examine the protein composition of plasma MPs of malaria samples in order to explore their effect on the pathogenesis of malaria and the possible linkage of circulating plasma MPs to the anaemia.

Methods

Aim

The aim of this study was to carry out proteomic analyses of MPs isolated from malaria positive samples and compare them with proteins of MPs from healthy controls in order to ascertain the role of MPs in malaria infection.

Design and setting

The study was a cross-sectional study conducted over a 2-year period, from May 2014 to June, 2016. The samples were collected at the Sunyani Regional Hospital, Ghana and all the laboratory work-up were carried out at the Department of Molecular Medicine, Kwame Nkrumah University of Science and Technology (KNUST), Noguchi Memorial Institute for Medical Research (NMIMR) and the Proteomic and Flow cytometry Core Facilities of the Indiana University School of Medicine.

The characteristics of participants/materials

In all, fifty three (53) participants who were out-patients were recruited into the study. There were 43 (23 males and 20 females) cases and 10 (5 males and 5 females) controls. The age range for controls was 28–47 years while the age range for patients was 1 week to 62 years.

Processes

Sample collection

Convenience sampling was employed for this study. Three milliliters (3 ml) of blood was collected from 43 laboratory diagnosed malaria patients. The parasite density was confirmed using independently prepared thick film. Control samples were obtained from 10 apparently healthy individuals. The samples were categorized into three (3): mild, moderate and high based on the level of parasitaemia.

Thick film preparation

Briefly, 6 microlitres (6 μl) of blood was placed in the middle of a glass slide for a thick film preparation using the WHO standardized template. The slides were then air-dried and subsequently stained as below.

Thick film staining and examination

The films were stained with 1:10 diluted Paskem® Giemsa for 10 min, air dried and examined using the Olympus ®CX 31 microscope (Tokyo, Japan). Parasite count was calculated using the formula

Where 6000 is the assumed WBC count per 1 μl of blood.

Purification of MPs from plasma

Briefly, EDTA-anticoagulated blood (3 mL) was centrifuged at 160 × g for 5 min, and plasma was stored at −20 °C until further processing. During processing, samples were thawed and centrifuged at 4000 x g for 60 min to obtain platelet-free plasma. The resultant supernatant was then spun for 120 min at 19,000 × g to obtain the microparticle pellet after the supernatant had been discarded [18]. The MP pellet was then dissolved in 50 μl of PBS and stored at −80 °C.

Flow cytometry analysis

All reagents used in the flow cytometry experiments were from BD Biosciences unless otherwise stated. 50 μl of phosphate buffered saline (PBS) was added to the thawed MP pellet. Equal volume of Annexin V binding buffer was added to label the MPs. Labeled plasma samples were analyzed on a BD FACSAriaTM flow cytometer. MPs isolated from plasma were gated (Annexin V+) based on their forward (FSC) and side (SSC) scatter distribution as compared to the distribution of synthetic 0.7–0.9 μm SPHERO™ Amino Fluorescent Particles (Spherotech Inc. Libertyville, Illinois, US) (Fig. ). Taking into account the presence of phophatidylserine (PS) residues in MPs surface, events present in Annexin V+ region were accessed for their positive staining with annexin V (BD Bioscience). The flow cytometry analysis done in this study was to verify the presence of Annexin V+ MPs in the pellet.

Fig 1

The figure shows a Venn diagram depicting overlap in MP proteins identified in Malaria, Culture Supernatant and Control samples

Proteomic analysis

One hundred microlitres (100 μl) of frozen MPs fractions were thawed and resuspended in 60% methanol and 0.1 M ammonium bicarbonate pH 8.0. Incubation with diothiothreitol followed by cysteine alkylation with iodoacetamide reduced the proteins. The solution was incubated overnight with mass spectrometry grade trypsin (Pierce, USA) at 37 °C to digest the proteins. The obtained tryptic peptides underwent desalting using reversed phase cartridge by washing in 0.5% acetic acid. The peptides were then eluted with 95% acetonitrile and 0.5% acetic acid. The eluted peptides were lyophilized and re-suspended in 0.5% acetic acid and directly analyzed using 2-dimensional liquid chromatography tandem mass spectrometry (2D-LC-MS/MS). The samples were loaded onto a microcapillary strong cation exchange column and fractions were collected using an increasing salt step elution gradient. Afterwards, each fraction was analyzed using the LTQ ion trap mass spectrometer (Thermo-Fisher Scientific, Waltham, MA). The acquired MS/MS spectra were searched against protein database available at www.uniprot.org [19]. The protein interactions were acquired using the string database available at http://string-db.org/cgi/.

Western blotting

Experimental Procedures: MPs were thawed and lyzed in sample buffer and were loaded on SDS-PAGE gels. The presence of the identified proteins in the samples was analyzed by Western blot. Bands at the predicted molecular weight for the proteins were observed [20].

All experiments (both independent and technical) were repeated 2–4 times.

Data analysis

Proteins identified were analyzed under molecular function, biological processes, subcellular location and cellular component using the uniprot database. Resultant data were further analyzed using SPSS 21.0 statistical software. Differences between the medians of the various groups were analyzed by Kruska Willis test. Pairwise correlations were evaluated with Spearman’s correlation coefficient r. A p-value < 0.05 was considered to be statistically significant. For illustration purposes the samples were categorized into 3 sub-groups according to the parasitaemia based on the old system of classification (1+, 2+ and 3+ which corresponded to parasite counts <5000/μl, >5000/μl but <10,000/μl and parasite count > 10,000/μl respectively in this study) to analyze their association with proteins released in MPs.

Results

Participant characteristics

The characteristics of subjects of this study are presented in Table . All patients were reporting on an outpatient basis. The age range for controls was 28–47 years while the age range for patients was 1 week to 62 years.

Table 1

Patient Characteristics

	Control	Patients
Number	10	43
Female (%)	50	47.5
Male (%)	50	53.5
Median Age (years)	37.5 (28–47)	22 (0.003–62)
Number below 5 years	0	9
Number pregnant	0	4

Plasmodium (malaria) species

The predominant plasmodium specie found in Ghana is falciparum. This phenomenon was seen in our study where all the patients tested had P. falciparum malaria. Subsequently the MPs types seen also carried P. falciparum markers as well as those of other activated cells (endothelial cells, erythrocytes, leukocytes and platelets).

Red blood cell indices

The red blood cell indices showed significant variation between the patients and the control. For example there was significance in haemoglobin concentration, MCV, MCH and RDW as seen in Table .

Table 2

Red Blood Cell Indices

Parameter (Unit)	Control	Patients	P-value
	Mean ± SD	Mean ± SD
Hb (g/dl)	13.20 ± 0.94	10.35 ± 2.61	0.037
Hct (%)	36.30 ± 3.35	30.74 ± 8.10	0.184
MCV (fl)	92.95 ± 2.37	80.68 ± 10.75	0.029
MCH (pg)	33.78 ± 0.35	26.82 ± 3.66	0.000
MCHC (g/dl)	35.75 ± 0.64	33.28 ± 2.64	0.072
RBC count (×1012/l)	3.83 ± 0.26	3.86 ± 0.83	0.934
RDW (%)	12.05 ± 0.29	15.80 ± 3.35	0.032

A p-value of < 0.05 was deemed to be statistically significant

Proteomics analysis

A total of 1729 proteins were identified in the malaria positive samples. The culture supernatant (CSN) showed 333 proteins while a total of 234 proteins were identified in the control samples. Seventy-two (72) proteins were common to the malaria positive samples, CSN and the control. The details are shown in Fig.  below.

The mean number of proteins obtained from MPs in malaria samples was 517.58 ± 56.58 while that of control was 191.50 ± 5.20. The independent sample t-test showed significantly higher number of proteins released in MPs from malaria samples. Twenty-nine P. falciparum proteins were identified in at least one malaria positive sample (Table ).

Table 3

P. falciparum proteins identified in MPs from malaria positive samples

Accession	Protein Name	Protein MW	Species
P04934	Merozoite surface protein 1	196199.8	PLAFC
P08569	Merozoite surface protein 1	193722.4	PLAFM
P13819	Merozoite surface protein 1	193721.5	PLAFF
P19598	Merozoite surface protein 1	192465	PLAF3
P86287	Actin-1	41871	PLAFX
Q8I4X0	Actin-1	41871	PLAF7
P10988	Actin-1	41843	PLAFO
P11144	Heat shock 70 kDa protein	74288	PLAFA
Q00080	Elongation factor 1-alpha	49041.2	PLAFK
P06719	Knob-associated histidine-rich protein	71941.5	PLAFN
P14643	Tubulin beta chain	49751.4	PLAFK
Q7KQL5	Tubulin beta chain	49751.4	PLAF7
P38545	GTP-binding nuclear protein Ran	24875.5	PLAFA
Q27727	Enolase	48704.2	PLAFA
Q8IJN7	Enolase	48678.1	PLAF7
Q9UAL5	Enolase	48662.1	PLAFG
P14140	Tubulin beta chain	49814.4	PLAFA
P13830	Ring-infected erythrocyte surface antigen	124907.8	PLAFF
Q25761	ADP-ribosylation factor 1	20840	PLAFO
Q7KQL3	ADP-ribosylation factor 1	20912	PLAF7
Q94650	ADP-ribosylation factor 1	20912	PLAFA
P12078	Heat shock 70 kDa protein PPF203 (Fragment)	23057.9	PLAFA
P13816	Glutamic acid-rich protein	80551.3	PLAFF
P19260	Merozoite surface antigen 2, allelic form 2	28555.5	PLAFG
P19599	Merozoite surface antigen 2	27890.3	PLAFF
P50490	Apical membrane antigen 1	71968.2	PLAFG
Q03498	V-type proton ATPase catalytic subunit A	68577	PLAFA
P06916	300 kDa antigen AG231 (Fragment)	33968.1	PLAFF
P04928	S-antigen protein	33695.1	PLAFN

Twenty three (23) RAB proteins were identified in MPs from malaria samples exclusively without any being found in the control samples as presented in Table . Out of the 1729 proteins found in the MPs from malaria positive samples, 653 were identified in plasma MPs from 1 category of samples only while 1076 proteins were identified in at least two categories and 697 proteins were found to be common to all 3 categories. The details of the relationship between the proteins released in plasma MPs from the 3 categories of malaria positive samples are shown in Fig.  below.

Table 4

List of Rab Proteins identified in MPs isolated from malaria samples

Accession	Protein Name	Protein MW	Species
Q9H0U4	Ras-related protein Rab-1B	22171.4	HUMAN
O00194	Ras-related protein Rab-27B	24608.1	HUMAN
P62820	Ras-related protein Rab-1A	22678	HUMAN
P61006	Ras-related protein Rab-8A	23668.4	HUMAN
P61106	Ras-related protein Rab-14	23897.2	HUMAN
P61026	Ras-related protein Rab-10	22541.1	HUMAN
P31150	Rab GDP dissociation inhibitor alpha	50583.2	HUMAN
P51149	Ras-related protein Rab-7a	23490	HUMAN
Q92930	Ras-related protein Rab-8B	23584.3	HUMAN
Q9NRW1	Ras-related protein Rab-6B	23461.9	HUMAN
P62491	Ras-related protein Rab-11A	24393.7	HUMAN
P50395	Rab GDP dissociation inhibitor beta	50663.7	HUMAN
Q15286	Ras-related protein Rab-35	23025.4	HUMAN
P20340	Ras-related protein Rab-6A	23593	HUMAN
P51148	Ras-related protein Rab-5C	23482.8	HUMAN
P51153	Ras-related protein Rab-13	22774.3	HUMAN
Q13637	Ras-related protein Rab-32	24997.5	HUMAN
Q96AX2	Ras-related protein Rab-37	24815.4	HUMAN
Q96E17	Ras-related protein Rab-3C	25952.4	HUMAN
Q9UL25	Ras-related protein Rab-21	24347.8	HUMAN
P61020	Ras-related protein Rab-5B	23707	HUMAN
P61019	Ras-related protein Rab-2A	23545.8	HUMAN
Q9NP72	Ras-related protein Rab-18	22977.3	HUMAN

Fig. 2

A Venn diagram depicting overlap of MP proteins identified in the 3 categories of malaria samples

Plasma MPs and inflammation

Some of the proteins identified showed strong link to inflammation. Inflammatory proteins seen in malaria MPs but not control sample MPs include Heat shock protein (HSP) 90-alpha, HSP 90-beta, 60 kDa HSP mitochondrial, 70 kDa HSP, HSP beta-1, Transforming growth factor beta-1 induced transcript 1 protein, Macrophage migration inhibitory factor and Transforming growth factor beta-1. Again, inflammatory proteins released in malaria MPs but in reduced quantities included: Heat shock cognate 71 kDa protein, Heat shock 70 kDa protein 1A/1B, Heat shock 70 kDa protein 6, Heat shock 70 kDa protein 1-like.

Plasma MPs and the complement system

A good number of complement associated proteins were seen in the MPs from the malaria samples. It was observed that the levels of Complement C3 and Complement C4-B, Complement C5, Complement C9, Complement factor B and Complement C1q subcomponent subunit C were significantly elevated in malaria plasma MPs compared to controls (Table ).

Table 5

Analysis of complement proteins from samples

Parameter	Patients	Control	P-value
Non-Parametric	Median (Q1-Q3)	Median (Q1-Q3)
Complement C3	192.5 (181–204)	134 (88–183)	0.083
Complement C4-B	103 (96–110)	38 (23–60)	0.000
C4b-binding protein alpha chain	18.5 (18–19)	15 (10.25–19.75)	0.192
Complement C1r subcomponent	15.5 (12–19)	16.5 (6.75–24.25)	0.895
Complement C1s subcomponent	9 (5–13)	6 (4–11.75)	0.311
Complement C5	11 (9–13)	4 (2–8)	0.620
Complement C1q subcomponent subunit B	3.5 (3–4)	5 (3–7)	0.142
Complement component C9	23 (21–25)	11 (7–18)	0.000
Complement factor B	11.5 (8–15)	5 (3–9.75)	0.025
Complement factor H	12.5 (5–20)	5 (2–10)	0.150
Complement C1q subcomponent subunit A	2 (1–4)	3 (2–5)	0.315
Complement C1q subcomponent subunit C	4 (2–6)	2 (1–3)	0.035
Complement component C7	6 (4–8)	2 (1.75–4)	0.061
Parametric	Mean ± SD	Mean ± SD
Complement component C6	4.27 ± 0.43	1.33 ± 0.52	0.000

A p-value of < 0.05 was deemed to be statistically significant

Plasma MPs and haemostasis

The following proteins associated with coagulation were release in malaria plasma MPs but not in control plasma. They are; Thrombospondin-1, Thrombospondin-4, Coagulation factor V, Coagulation factor XII and Coagulation factor XIII A chain. Again, Fibrinogen beta chain and plasminogen released in the malaria plasma MPs were significantly higher compared to control plasma. There was however no significant difference in levels of Fibrinogen alpha chain, Fibrinogen gamma chain and von Willebrand factor between the 2 groups. Antithrombin-III in plasma MPs was however significantly reduced in the patients compared to the control group. The summary is presented in Table .

Table 6

Analysis of haemostatic proteins from samples

Parameter	Control	Patient	P-value
Non-Parametric	Median (Q1-Q3)	Median (Q1-Q3)
von Willebrand factor	11.5 (11–12)	15 (5.5–27)	0.493
Antithrombin-III	17 (13–21)	5.5 (3–8)	0.001
Thrombospondin-1	39 (17.25–71.25)	0	0.000
Coagulation factor V	16.50 (5.25–36.50)	0	0.000
Coagulation factor XIII A chain	8 (1–19)	0	0.000
Thrombospondin-4	0.00 (0–3.75)	0	0.043
Coagulation factor XII	1 (0–2)	0	0.004
Plasminogen	30.0 (10.25–43.75)	0	0.000
Parametric data	Mean ± SD	Mean ± SD
Fibrinogen alpha chain	88.5 ± 0.58	59.01 ± 28.35	0.045
Fibrinogen beta chain	74.13 ± 6.93	56.79 ± 27.62	0.225
Fibrinogen gamma chain	53.37 ± 9.24	37.56 ± 16.05	0.066

Plasma MPs and haemoglobin sub-units

Some haemoglobin subunit proteins were seen in the circulating MPs from both malaria positive samples and controls. The quantities of haemoglobin subunits in plasma MPs from malaria positive samples were however significantly higher as compared to controls (Table 7).

Table 7

Analysis of haemoglobin subunit proteins in samples

Parameter	Control	Patients	P-value
	Median (Q1-Q3)	Median (Q1-Q3)
Haemoglobin subunit gamma-2	1 (1–1)	23 (10–63)	0.001
Haemoglobin subunit beta	13 (11–15)	143 (69–274)	0.000
Haemoglobin subunit gamma-1	1 (1–1)	28 (10.5–64.5)	0.000
Haemoglobin subunit alpha	11 (9–13)	125 (52–284)	0.000
Haemoglobin subunit delta	3.5 (3–4)	72 (40–138)	0.000
Haemoglobin subunit epsilon	1 (1–1)	21 (8–51)	0.000

Plasma MPs and adhesive proteins

Some adhesive and receptor proteins were identified in the plasma MPs from the malaria samples but they were absent in the control samples. They include: Merozoite surface protein 1, Knob-associated histidine-rich protein, Disintegrin and metalloproteinase domain-containing protein-10,-12,-17, Complement receptor type 1, Cysteine and glycine-rich protein 1, Ring-infected erythrocyte surface antigen, Glycophorin-binding protein, Intercellular adhesion molecule 2, A disintegrin and metalloproteinase with thrombospondin motif 13, Merozoite surface antigen 2, Endothelial cell-selective adhesion molecule.

Plasma MPs and cytoskeletal proteins

Cytoskeletal proteins were identified in the malaria samples but they were absent in the control samples. Spectrin alpha chain, erythrocytic 1, Spectrin beta chain erythrocytic 1, Coronin 1A, Coronin 1C, Myosin 9, Actin-1, Actin-related protein 3, Actin-related protein 2, Actin-related protein 2/3 complex subunit 3, Actin-related protein 2/3 complex subunit 4 and Actin-related protein 2/3 complex subunit 5.

Flow cytometer distribution for MPs show a typical forward-/side-scatter distribution

MPs can be identified by their forward-/side- scatter appearance on flow cytometer. As can be seen from Fig. , the forward/side scatter distribution of the MPs from the samples shows a similar pattern in line with the classical appearance of MPs on flow cytometer and comparable to those obtained by others in similar experiments. This, in combination with other properties such as the proteins detected give credence to the fact that throughout our experiments, MPs were being obtained.

Fig. 3

Read out from flow cytometry depicting the classical appearance of MPs

Protein enrichment of the isolated MPs

Figure 4 shows protein enrichment of MPs (A) and proteins differentially expressed by MPs (B) by Western blot. The proteins were loaded according to the sample categories of mild, moderate and high malaria as well as no-malaria and bands at the predicted molecular weight for each of the proteins were observed.

Fig. 4

A figure showing protein enrichment of MPs a and proteins differentially expressed by MPs b through western blotting. The proteins were loaded according to the sample categories of mild, moderate and high malaria as well as no malaria and bands at the predicted molecular weight for each of the proteins were observed. The figure shows representative gels of 2–4 experiments

Correlation coefficient of clinical variables

Tables , and show the Spearman correlation coefficients of various groups of proteins identified in MPs from malaria samples against the parasite count and peripheral haemoglobin concentration. There was no significant correlation of all the haemoglobin subunits against either the peripheral haemoglobin concentration or parasite count. Peripheral haemoglobin however negatively correlated with parasite count (r = −0.415, p < 0.001) and Mannose-binding lectin serine protease 2 (r = −0.664, p < 0.001) while it positively correlated with Complement C3 (r = 0.457, p < 0.001). Parasite count correlated negatively with Complement C3 (r = −0.359, p < 0.05) and correlated positively with Complement C1s subcomponent (r = 0.407, p < 0.001) and C4b-binding protein beta chain (r = 0.845, p < 0.05).

Table 8

Spearman’s rank correlation coefficients of Haemoglobin subunits and coagulation proteins released in MPs against peripheral haemoglobin concentration and parasite count

Microparticle Protein	HB	Parasite Count
HB		−0.415**
Haemoglobin subunit epsilon	0.2	−0.25
Haemoglobin subunit gamma-2	0.201	−0.227
Haemoglobin subunit beta	−0.063	0.085
Haemoglobin subunit gamma-1	0.197	−0.17
Haemoglobin subunit alpha	−0.09	0.212
Haemoglobin subunit delta	−0.059	0.13
Haemoglobin subunit mu	−0.5	0.5
Alpha-Haemoglobin-stabilizing protein	0.5	−0.5
Fibrinogen beta chain	0.211	−0.254
Fibrinogen alpha chain	0.029	0.121
Fibrinogen gamma chain	0.285	−0.131
Coagulation factor V	0.177	−0.151
Coagulation factor XIII A chain	0.053	0.037
von Willebrand factor	0.036	0.047*
Antithrombin-III	0.175	−0.079
Coagulation factor XII	0.2	−0.006

*Correlation significant at p < 0.05

**Correlation significant at p < 0.001

Table 9

Spearman’s rank correlation coefficients of Complement proteins released in MPs and peripheral haemoglobin concentration and parasite count

Complement Proteins	HB	Parasite Count
Complement C4-B	0.21	−0.161
Complement component C6	−0.207	−0.414
Complement C3	0.457**	−0.359*
C4b-binding protein alpha chain	0.095	0.103
Complement C1r subcomponent	0.006	0.045
Complement C1s subcomponent	−0.081	0.407**
Complement C5	−0.017	0.025
Complement C1q subcomponent subunit B	0.147	−0.191
Complement component C9	0.045	−0.111
Complement factor B	0.007	0.144
Complement factor H	0.078	−0.004
Complement C1q subcomponent subunit A	−0.071	0.352
Complement C1q subcomponent subunit C	0.124	−0.224
Complement component C8 beta chain	0.202	0.251
Complement component C7	0.025	0.350
Complement factor I	−0.059	−0.736
Complement component C8 alpha chain	0.421	0.089
Complement component C8 gamma chain	−0.344	0.465
Complement decay-accelerating factor	−0.514	0.062
C4b-binding protein beta chain	−0.338	0.845*
Complement factor H-related protein 2	−0.258	0.775
Complement factor I	−0.059	−0.736
Complement component C8 alpha chain	0.421	0.089
Complement component C8 gamma chain	−0.344	0.465

*Correlation significant at p < 0.05

**Correlation significant at p < 0.001

Table 10

Spearman’s rank correlation coefficients of selected microparticle proteins against Parasite count and Haemoglobin

Microparticle Protein	HB	Parasite Count
Leukocyte elastase inhibitor	−0.106	−0.206
P-selectin	0.078	−0.023
Mannan-binding lectin serine protease 2	−0.664**	0.176
Intercellular adhesion molecule 1	−0.686	0.169
Intercellular adhesion molecule 2	−0.093	−0.109
Intercellular adhesion molecule 3	−0.726*	0.676
Mannan-binding lectin serine protease 1	−0.157	0.258

*Correlation significant at p < 0.05

**Correlation significant at p < 0.0

Discussion

The study identified an array of proteins from the MPs isolated from the malaria positive samples, culture supernatant and healthy controls. The mean haemoglobin (Hb) of the studied subjects was significantly lower than that of the control group (p = 0.037). This finding agrees with a study in the Brazilian Amazon in which out of a total of 7831 people studied, individuals with malaria were seen to have the lowest Hb compared to community controls and malaria-negative febrile patients [21]. The results of our study also collaborate with the results from a study carried out in Nigeria [22]. These observations are expected because, as far back as in 1947 the plasmodium was reported to “consume” haemoglobin during its growth within erythrocytes [23]. P. falciparum for instance is capable of digesting more than 80% of the erythrocyte haemoglobin [24]. This goes to explain the significant negative correlation of Hb and parasite count (r = −0.415, p < 0.001) indicating that the greater the parasite density in a patient the lower the Hb, as was also reported in a study in Thailand [25].

Again, there was a reduction of the haematocrit (Hct) of the patients (Table ) as compare to controls which was however not significant. Similar results were reported in patients with uncomplicated malaria where the authors indicated that, uncomplicated malaria was associated with milder biochemical alteration and haemolysis as opposed to complicated/severe malaria [26]. A study done at ECWA Community Health Centre, Bukuru, Jos in Nigeria however showed significant reduction of the haematocrit of malaria samples compared to healthy controls [22].

The mean cell volume (MCV) of the patients was significantly lower than that of the control group (Table ). This result was unexpected because previous studies in a population near Thailand-Myanmar border indicated the opposite finding usually because as erythrocytes are being destroyed in malaria infection, the bone marrow is stimulated to push more new erythrocytes whose mean cell volume are higher than older erythrocytes [25]. Also, the mean cell haemoglobin (MCH) of patients was significantly lower than that of the control group. This result however contradicts with the results of the population near the Thailand-Myanmar border where the predominant malaria parasite is P. vivax [25]. Therefore the results seen in this study can probably be explained by the ability of P. falciparum (predominant in Ghana) to consume haemoglobin during its growth in the erythrocyte [23] and the propensity of the erythrocytes in malaria to release more haemoglobin-bearing MPs into circulation leaving the erythrocytes with less haemoglobin compared with erythrocytes from healthy controls [23].

The red cell distribution width (RDW) was significantly higher (p = 0.032) in the patients compared to the control group. It was however consistent with results from a P. vivax malaria study [27]. The mean RDW was 15.80 and RDW greater than 15 has been shown to be a predictive value of malaria infection, although some researchers contest this [28, 29]. In P.vivax malaria however, the elevated RDW results from the initial size increase of parasitized erythrocytes which is followed by erythrocyte rapture. The same mechanism however cannot be said to be associated with RDW increase in P. falciparum infection since parasitized cells maintain their sizes [29]. The elevated RDW seen in this study could result from the bone marrow’s effort to balance erythrocytes loss in malaria by pushing more new erythrocytes into circulation which also increases the macrocyte percentage [30, 31].

Again, our data indicated that a significantly higher amount of all haemoglobin subunits released came from MPs of malaria positive samples. This point to an interesting phenomenon where the erythrocytes from malaria patients lose their haemoglobin to vesiculation and the vesicles that result may be eliminated by the body thereby contributing to anaemia. Table  shows a list of 24 RAB proteins found in the malaria samples. They include proteins which play pivotal roles in relation to MP docking, fusion and appropriate targeting of various cellular compartments [32].

Also, among the proteomics data were some erythrocyte cytoskeletal proteins which have been referred to as low abundance cytoskeletal proteins [33]. They include myosin chains, moesin, ezrin and F-actin capping proteins. Furthermore, some Pf proteins which have been described as potential drug targets were identified in the MPs, they include: enolase, Hsp 90, hypoxanthine guanine phosphoribosyl transferase, L-lactate dehydrogenase and Phosphoglycerate kinase. This is because the parasite through these proteins in their pathways starts processes such as glycolysis, haemoglobin digestion and salvage of purines [34]. Tubulin Beta which is a brain antigen that can discriminate cerebral malaria from other forms of the disease was identified in MPs from 38 samples [35]. This protein chain is an antimalarial target [36].

Enolase is a glycolytic enzyme and especially essential in ATP generation in organisms that are devoid of the Krebs cycle. In malaria, surface enolase assists in parasite invasion by binding to plasminogen [37]. Enolase was found in MPs of 38 of the samples. It is known to play a crucial role in the parasite invasion of the midgut of the mosquito [38]. A study has shown that its levels are elevated in infected red blood cells compared to uninfected cells. Antibodies against merozoite surface enolase have the ability to interfere with P. falciparum invasion of the red cell thereby conferring a partial protection against malaria [39]. Knob-associated histidine-rich protein (KAHRP) was identified in only 1 of the 43 malaria samples. KAHRP is a major component of the knob which is an electron-dense protrusion located on the membrane of infected erythrocytes. The knob is the site of adhesive interaction between infected erythrocytes and vascular endothelial cells [40].

Plasmodium falciparum Merozoite surface protein 1 (MSP-1) which is also called precursor to major merozoite surface antigens (PMMSA) or merozoite surface antigen 1 (MSA-1) was identified in MPs from 2 of the malaria samples [41]. This protein is the most widely studied of proteins of Plasmodium falciparum. MSP-1 is well conserved among P. falciparum isolates and hence it was identified in the MP isolated from the culture supernatant [42]. This protein binds to erythrocytes in a sialic dependent manner suggesting it is a receptor to a ligand on erythrocyte surface permitting adhesion of Plasmodium falciparum [41].

Also, merozoite surface antigen 2 (MSP2) which is a surface coat protein essential for the survival of the blood-stages of P. falciparum was identified [43]. This protein and its allelic form 2 were identified in MPs from one sample. Also identified in one sample was Plasmodium falciparum Apical membrane antigen 1 (PfAMA1) which is synthesized during schizogony and transported to micronemes. PfAMA1 translocates onto the merozoite surface prior to erythrocyte invasion when it serves as an adhesion molecule playing a central role in the invasion process [44]. Ring-infected erythrocyte surface antigen (RESA), a protein expressed in early stage gametocytes, final stages of schizont and stored in dense granules within the merozoites was found in one sample. RESA binds to spectrin, its primary site on erythrocyte and is associated with the membrane of recently invaded erythrocyte but it is only evident in the cell up to 24 h after parasite invasion [45].

Glutamic acid-rich protein (GLURP) found in one sample is a molecule located on the surface of merozoites and also in the hepatic stage [46]. This protein is an antigen and its epitopes defined by non-repetitive sequence are thought of to be more effective antibody-dependent cellular inhibition process and is believed to be involved in acquired protective immunity to malaria. These epitopes are therefore proposed to be preferred in vaccine formulations against malaria [47, 48]. Also, P. falciparum actin 1(PfACT1) was expressed in the MPs of 35 samples. PfACT1 is a ubiquitously expressed protein and it is expressed throughout the lifecycle of the Plasmodium. However, its isoform PfACT 2 which is expressed only in the sexual stages of the parasite was not isolated in any of the samples [49].

Heat shock 70 kDa protein (Hsp70) was expressed in 35 samples. P. falciparum has six Hsp70s [50]. Hsp70 proteins have molecular chaperone functions, and are involved in a number of processes including protein degradation and controlling of the activity of regulatory proteins, protein folding and protein translocation across membranes. Hsp 70 members are present in almost all cellular compartments [51] and this could account for its identification in a good number of our samples. Also, elongation factor-1 alpha was found in the MPs from 7 of the malaria positive samples. It is an abundant protein that is an essential element in eukaryotic protein translation, in species of Plasmodium [52]. It is key in the proliferation of the blood stages of Plasmodium [53].

Precursors of serine-repeat antigen (SERA) proteins are synthesized in the late trophozoites [54]. Among organisms in the apicomplexan phylum and with the exception of Theileria found in cattle, species of Plasmodium are the only organisms that the gene family translating into Serine-repeat antigen (SERA) protein has been found [55]. This protein was found in 1 sample. Investigation with anti-SERA antibody has established SERA protein as a target for antibody dependent cellular inhibition of P. falciparum development [56].

Also identified was Plasmodium falciparum ADP-ribosylation factor 1 (PfARF1) which is activated after binding to GTP. In the secretary pathway, PfARF1 regulates vesicular biogenesis and trafficking processes [57]. It is also thought of to be involved in the transport of MSP-1 from the endoplasmic reticulum and plays a role in the activation of a calcium-signalling mechanism in the parasite [58]. Hypothetical protein identified 300 kDa antigen AG231 (Fragment) was also found in 1 sample. This protein was discovered in about 93% of 65 patients living in a malaria endemic area in Papua Guinea. It is found in schizonts and trophozoites. Again, P. falciparum S-antigen proteins were found in the MPs of 2 samples. They are heat-stable antigens that are serologically diverse among varied isolates of the parasite [59].

There was significant difference in the levels of the following complement system proteins released in plasma circulating MPs between the malaria samples and that of control: complement component C6, complement C4-B, complement component C9, complement factor B and Complement C1q subcomponent subunit C. The difference between the patients and the control group in complement component C3 was however not significant but correlation analysis indicated a significant negative correlation between C3 released in MPs and parasite count in the malaria samples. There was significant positive correlation between C3 and haemoglobin (r = 0.457). It is known that MPs are able to bind to complement component C1q and cause the C3 to be fixed on the MPs after exposure to normal human serum through the classical pathway. These complement-fixed MPs are capable of binding to erythrocytes and remove them from circulation [60]. It could therefore be presumed that complement-fixed MPs bound to erythrocyte might be a mechanism that explains the increase erythrocyte clearance in malaria leading to reduced haemoglobin concentration. This presumption can be related to a similar work that explains the role of MPs in complement activation relating to the pathogenesis of rheumatoid arthritis [61].

Also, Hb negatively correlated with Mannan-binding lectin serine protease 2 released in MPs (Table ). This may be explained by the ability of this protein to activate complement and interact with phagocytes and also serves as an opsonin to Plasmodium infected erythrocytes facilitating the elimination of such erythrocytes from circulation and invariably causing reduction in Hb [62].

C4 binding protein beta chain in MPs from malaria samples correlated positively with parasite count. C4 binding protein is an abundant plasma protein whose natural function is to inhibit the classical and lectin pathways of complement activation. Research has shown that the fusion of oligomerization domains of C4 binding protein to MSP-1 from Plasmodium yoelii improved its immunogenicity [63]. C4 binding protein beta chain released in MPs could hence be acting as an adjuvant to MSP-1 antigens thereby eliciting the appropriate immune response against the parasite. Antithrombin concentration with the MPs in the patients was significantly low compared with the controls [64]. Serum concentration of antithrombin III particularly in severe malaria is reduced [65, 66]. This is because generally malaria is associated with the consumption of Antithrombin [67]. Figures 5 and 6 represent isolated proteins that are involved in biological and molecular functions, Fig. 7 represent isolated proteins that are cellular components and Fig. 8 a network model delineating the P. falciparum proteins isolated from MPs of malaria positive samples and their associated pathways.

Fig. 5

3D pie chart of the Gene Ontology of all P. falciparum proteins isolated from samples that are involved in biological processes

Fig. 6

3D pie chart of Gene Ontology of the P. falciparum proteins isolated from the malaria samples that are involved in molecular functions

Fig. 7

3D pie chart of Gene Ontology of P. falciparum proteins isolated in the malaria samples that are cellular components

Fig. 8

A network model delineating the P. falciparum proteins isolated from the MPs in malaria positive plasma and their associated pathways. The node colours represent MP proteins and their interactions (colour coded basis for a particular interaction)

In terms of limitations, because convenience sampling was used in this study further stratification of the patients and control was not done to assess the association of parameters like age category, sex and duration of onset of symptoms on the proteomic profile of subjects. If possible, a further prospective sampling involving sex and age matched controls should be employed in future study so that data could be stratified to address the afore-mentioned limitation.

Conclusion

Our study has clearly established that there is an up regulation of protein content of circulating plasma MPs from malaria patients than in healthy controls. A great variety of protein content can be seen in malaria patients although a large number of proteins identified were common to the 3 categories of malaria samples. Furthermore, higher amounts of haemoglobin subunits are released into circulating MPs of malaria patients. These MPs when finally eliminated from circulation have the propensity to cause anaemia. Interaction enrichment indicates that the P. falciparum proteins isolated from the malaria samples are least partially connected as a group.

The protein-protein interaction (PPI) of the interaction of the P. falciparum indicated that the proteins have more interactions among themselves than what is expected for a random set of proteins of similar size drawn from the P. falciparum genome. The enrichment indicates the proteins are at least partially connected as a group.