A Consequence of Inflammatory Cytokine

7/31/2019 A Consequence of Inflammatory Cytokine

1/32

BioMedCentral

Page 1 of 32(page number not for citation purposes)

Malaria Journal

Open AccesReview

Human malarial disease: a consequence of inflammatory cytokinerelease

Ian A Clark*1

, Alison C Budd1

, Lisa M Alleva1

and William B Cowden2

Address: 1School of Biochemistry and Molecular Biology, Australian National University, Canberra, ACT 0200, Australia and 2John Curtin Schoolof Medical Research, Australian National University, Canberra, ACT 0200, Australia

Email: Ian A Clark* - [email protected]; Alison C Budd - [email protected]; Lisa M Alleva - [email protected];William B Cowden - [email protected]

* Corresponding author

Abstract

Malaria causes an acute systemic human disease that bears many similarities, both clinically and

mechanistically, to those caused by bacteria, rickettsia, and viruses. Over the past few decades, a

literature has emerged that argues for most of the pathology seen in all of these infectious diseases

being explained by activation of the inflammatory system, with the balance between the pro and

anti-inflammatory cytokines being tipped towards the onset of systemic inflammation. Although notoften expressed in energy terms, there is, when reduced to biochemical essentials, wide agreement

that infection with falciparum malaria is often fatal because mitochondria are unable to generateenough ATP to maintain normal cellular function. Most, however, would contend that this largely

occurs because sequestered parasitized red cells prevent sufficient oxygen getting to where it is

needed. This review considers the evidence that an equally or more important way ATP deficency

arises in malaria, as well as these other infectious diseases, is an inability of mitochondria, throughthe effects of inflammatory cytokines on their function, to utilise available oxygen. This activity of

these cytokines, plus their capacity to control the pathways through which oxygen supply to

mitochondria are restricted (particularly through directing sequestration and driving anaemia),

combine to make falciparum malaria primarily an inflammatory cytokine-driven disease.

BackgroundThe mechanism of the disease caused byPlasmodium falci-parum, arguably the pathogen that causes the most humansuffering, has been hotly debated for many decades.Clearly, rational adjunct therapy depends on getting thisright. For over twenty years the central debate has comefrom two apparently opposing camps. One championsthe mechanical hypothesis, based on the concept of insuf-ficient oxygen reaching vital organs, and the other thecytokine hypothesis, in which excessive release of pro-inflammatory cytokines are the primary driving force ofdisease and death. The former concept stresses the

uniqueness of the pathophysiology of falciparum malaria

compared to that of other severe systemic infectious dis-eases, whereas the latter sees malaria as having fundamen-tally the same basis as these other conditions, with theadhesive property of parasitized erythrocytes giving it nomore than a distinctive flavour.

Critical analysis of the mechanism of falciparum malarialdisease would not have been possible without the seminal

work of Peter Mitchell [1,2], who identified mitochondriaas the ATP-generating powerhouse of aerobic cells, andthus of all aerobic organisms. Among the doors this

Published: 10 October 2006

Malaria Journal2006, 5:85 doi:10.1186/1475-2875-5-85

Received: 25 August 2006Accepted: 10 October 2006

This article is available from: http://www.malariajournal.com/content/5/1/85

2006 Clark et al; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0),which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
http://www.biomedcentral.com/http://www.biomedcentral.com/http://www.biomedcentral.com/http://www.biomedcentral.com/http://www.biomedcentral.com/info/about/charter/http://-/?-http://-/?-http://www.malariajournal.com/content/5/1/85http://creativecommons.org/licenses/by/2.0http://www.biomedcentral.com/info/about/charter/http://www.biomedcentral.com/http://-/?-http://-/?-http://creativecommons.org/licenses/by/2.0http://www.malariajournal.com/content/5/1/85


2/32

Malaria Journal2006, 5:85 http://www.malariajournal.com/content/5/1/85


opened was the opportunity to understand severe infec-tious disease by seeing it through the perspective of theseorganelles. This review largely discusses the relative contri-bution to disease processes of mitochondria being pre-

vented from getting enough oxygen, and being able to use

all the oxygen that reaches them, and the combined effectsof these two influences. This includes blood flow restric-tion in microvessels by parasitized red cells that adhere toendothelial walls (sequestration), the potential restrictionto oxygen supply that is unique to falciparum malaria. Asdiscussed below, the broader literature is consistent withthe view that sequestration during severe disease is not apassive event that is simply an amplification of whatoccurs early in infection and in tolerant individuals, butone in which location and avidity of adherence, and,therefore, pathogenic effects, are controlled by inflamma-tory cytokines.

Despite advances in understanding diseases clinically verysimilar to falciparum malaria in terms of inflammatorycytokines, enthusiasm for the mechanical obstructionhypothesis seems at least as strong as ever [3-5]. Althoughmuch literature now demonstrates dependence of themechanisms of poor oxygen delivery on excess inflamma-tory cytokine release, even the most recent members of the

vaso-occlusion school [5] still see the cytokine theory ofdisease as an alternative to be argued against rather thanas an essential component of their own disease model.Hence it is timely to bring together an update of the evi-dence why cytokines are regarded as so central to this dis-ease. In particular, it seems warranted to summarize how

broadly the harmful influences of the inflammatorycytokines are now known to extend. By introducing con-cepts in infectious disease in general, and then moving tothe particular case of falciparum malaria, this reviewexpands on these functionally interconnected conse-quences of excess production of inflammatory cytokines.

Systemic infectious diseases and inflammatorycytokines

There is now remarkably widespread acceptance thatcytokines such as TNF and interleukin-1 (see "cytokinestorm" in Google) are the essential mechanism of sys-temic disease caused by infectious agents. Indeed, one

would be hard pressed to find an alternative explanationnow current for the anorexia, tiredness, aching joints andmuscles, fever and sleepiness that patients experience inany systemic infection, including both vivax and falci-parum malaria. Neither is it disputed that exacerbatedrelease of these same mediators is the best current line ofinvestigation for the mechanism of severe and life-threat-ening illness, such as sepsis [6] and influenza [7]. The dif-ficulty, confined to falciparum malaria, is to get its broadacceptance in a research community that has traditionallyseen its disease as unique, mechanistically separated from

other infectious conditions by the presence of sequesteredparasitized red cells often seen in certain intravascularlocations at autopsy. To some researchers sequestered par-asites are still necessary and sufficient for illness from fal-ciparum malaria to occur, and to cause fatality [4,8].

Much of this section will include parallels betweenmalaria and similar diseases caused by other pathogens,and basic research done on the effects of inflammatorycytokines on normal cells. It also recounts, for the present-day audience, the malarial origin of this concept of diseasepathogenesis.

Intra-erythrocytic death of haemoprotozoa, the original

link of TNF to disease

Nearly thirty years ago, ignorance of accepted malaria wis-dom, in a London tumour/virology lab where thisattribute was universal, allowed a novel tumour-orientedinterpretation to be put on the observation that the non-

lethal mouse parasite, Babesia microti, was killed by theimmune system in circulating red cells [9]. This also hap-pened in some species of malaria parasites with which itcross-protected. Encouragement that this phenomenon

was worth pursuing came from the Guy's Hospital group,then at the top of the malarial immunity tree, whoobserved the same unexpected and puzzling phenome-non when they challenged malaria-immunized rhesusmonkeys [10].

Of particular concern for the then official Guy's dogma(malarial immunity operates through a specific antibodyfocussed on the merozoite surface, adopted unquestioned

by other major vaccine groups), even unrelated parasitesdied en masse inside red cells when previously immunizedmonkeys were challenged [11]. In their view this couldoffer an explanation of why some monkeys had high lev-els of antibody expected to be protective, yet failed toresist a challenge infection [11], while others with little orno anti-merozoite antibody were immune [12]. A reportfrom the US told a similar tale [13]. Clearly, some power-ful influence beyond specific antibody, and inconvenientfor mainstream thinking on immunity and vaccine devel-opment, was reproducibly occurring.

Tumour necrosis factor (TNF), the prototype

inflammatory cytokineThe observation that pre-treatment with the Bacillus Cal-mette-Gurin (BCG) strain ofMycobacterium tuberculosiscontrolled a subsequent infection with any of severalstrains of babesia or malaria in mice (no antibody, para-sites dying in red cells, not phagocytes) was fortuitouslytimed with the publication of the first paper on TNF [14].

This allowed us, in collaboration with these New Yorktumour researchers, to propose novel roles for TNF inimmunity and disease pathogenesis in malaria and sepsis[15,16]. In summary, through linking the protective
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


3/32



capacity of agents such as BCG with the degree to whichthey sensitised to bacterial lipopolysaccharide (LPS) itcame to be realised that the pathology of LPS toxicity (andsubsequently that of rTNF) and Plasmodium vinckei infec-tions in mice were largely identical, cytokine-mediated,

events. As noted in a contribution to Brian Maegraith's1988 Festschrift Proceedings [17], the cytokine approachgave teeth to his inflammation-based arguments onmalaria disease of forty years earlier [18].

As reviewed in 1987 [19], this experience with mousebabesiosis and malaria provided the insight that this anti-tumour mediator arguably had roles in both cell-medi-ated immunity (CMI) and the pathogenesis of infectiousdisease in general. As well as malaria, this concept was rea-soned in 1981 to explain the mechanism of typhoid [15],sepsis in general [16], and viral diseases in 1989 [20], andit eventually spread across all acute infectious diseases

(see [21] for a recent review). For example, within a fewyears it began to dominate the sepsis literature [22,23],and the virulence of different strains of influenza, a dis-ease that is a standard clinical misdiagnosis for importedmalaria, has now been expressed in terms of their capacityto induce TNF [24]. While still engendering strong oppo-sition from some malaria researchers [4,25], these ideashave been readily accepted by scientists working on bacte-rial, rickettsial or viral diseases. A broad literature acrossinfectious disease now describes inflammatory cytokinesas having a beneficial role in host defence, but beingharmful to the host if produced excessively. Indeed, theacceptance and applicability of this concept is now gen-

eral enough for its biological evolution to be an inde-pendent subject for research [26].

Once neutralizing anti-TNF antibodies became availablefor human use, they were tested by others for efficacyagainst malarial parasites and disease. Unfortunately acentral tenet of the concept (that the pro-inflammatorycytokines that cause disease are the same mediators that,in lower concentrations, are responsible for the innateimmunity that controls parasite growth see also tubercu-losis etc. in next paragraph) was not adequately consid-ered. TNF has been shown to inhibit a mouse malariaparasite in vivo [27], and Plasmodium falciparum in vitro,

provided white cells to generate the next down-streammediator, possibly nitric oxide [28], were present [29].

This is consistent with findings in human subjects [30].Thus it is not surprising that anti-TNF antibody, by remov-ing inhibitory pressure from the pathogen, can enhancethe disease in falciparum malaria [31], as shown five yearsearlier in human sepsis [32].

The broad relevance of these malaria-origin concepts inimmunity and disease is best illustrated by noting the con-sequence of passive vaccination against TNF for Crohn's

disease and rheumatoid arthritis, now a large-scale rou-tine treatment [33]. Its practical success puts the relevanceof these pro-inflammatory cytokines in human inflamma-tory disease beyond doubt, and the major side effect (pre-existing or acquired tuberculosis, salmonellosis, or listeri-

osis becoming fulminant) nicely demonstrates its rele-vance to CMI against many pathogens. It is unlikely to becoincidental that all three that flare are on our list oforganisms that protected against haemoprotozoan para-sites, causing intra-erythrocytic death, and priming micefor TNF production [34,35]. Evidently the host was settingup a cell-mediated response that would protect againstthese organisms. Being non-specific in nature, it also pro-tected against haemoprotozoa as well. From this reason-ing Coxiella burnetii, a crude extract of which was anextremely good protectant [36], and primer for TNF (E.Carswell, pers. comm.) will also predictably flare if anti-

TNF is given, long term, to an arthritis patient harbouring

this human pathogen.

The inflammatory cytokines as a group

In this text, TNF is used as a term of convenience to desig-nate the pro-inflammatory cytokines as a whole. Othercytokines, such as lymphotoxin (LT), interleukin-1 (IL-1),interleukin-6 (IL-6) and soluble Fas ligand (FasL) servesimilar functions. In passing, it warrants noting that theterm TNF-alpha, while still common, has been obsoleteever since LT ceased being referred to as TNF-beta andreverted to its original name, allowing TNF to do thesame. Although the literature connecting the pro-inflam-matory cytokines other than TNF to malaria [37-40] is as

yet much smaller than that for TNF, this does not implythat their potential for understanding this disease is corre-spondingly minute. A TNF superfamily of 19 memberssignalling through 29 receptors has more recently beendescribed [41]. Many of these mediators induce othercytokines and enzymes that add to the inflammatory cas-cade. For example, TNF induces migration inhibitory fac-tor (MIF) [42,43], and TNF, IL-1 beta and LT generate theinducible form of nitric oxide synthase iNOS [44]. Anti-inflammatory cytokines such as IL-10, IL-4, and trans-forming growth factor-beta (TGF-beta), also play activeroles, and an imbalance between these and their pro-inflammatory counterparts often determines outcome in

disease. Some tens of thousands of publications oninflammatory cytokines and systemic inflammatory andother disease now exist.

The pro-inflammatory cytokines most closely investigatedin malaria, such as TNF, usually act as homeostatic agents,but can cause pathology if produced excessively. Arecently defined example is MIF (see above paragraph),belonging to an ancient gene family, with structuralhomologues in bacterial organisms. As an indication ofthe broad relevance and complexity of these cytokines and
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


4/32



their interactions, MIF is also induced in mammals byE.coli lipopolysaccharide [42], staphylococcus toxic shocktoxin, and streptococcal pyrogenic toxin A [45]. Whetherthis is direct, or via the TNF they induce, appears not tohave been ascertained. Likewise, MIF can act directly,

through TNF, or in synergy with it, in generating anaemia,as discussed below. It is, as noted above, TNF-induced,and remarkable for several reasons, one being that itsdescription 40 years ago [46,47], began the concept of

what are now called cytokines. Ten years later, and longbefore MIF was realised to have functions other thanmigration inhibition and pathogenicity in sepsis (whichits inhibition suppresses, in a realistic model, mostimpressively [48]), it was the first cytokine described in amalaria infection [49]. A few years after its rediscovery asa homeostatic glucocorticoid antagonist [50,51], it hasbecome central to understanding malarial anaemia, asdiscussed below. It is also increased in malarial placentas

[52].

Cytokines such as TNF and IL-1, both increased in a widerange of systemic inflammatory diseases, including falci-parum malaria, can induce a late-onset, but long-acting

wave of a cytokine termed the high mobility group box 1(HMGB1) protein, which prolongs and amplifies inflam-mation [53,54]. This molecule, previously known for sev-eral physiological functions, now shows great promise asa therapeutic target in sepsis, in that countering it after theonset of illness has been reported to protect well in exper-imental sepsis [55,56]. HMGB1 has been shown to beincreased, in proportion to degree of illness, in serum

from African children infected with falciparum malaria[57]. Like TNF, HMGB1 has roles in other inflammatorydiseases [58], reaffirming malaria's position within theirranks. The malarial context of HMGB1 is reviewed morefully elsewhere [21].

TNF, a tool to determine the nature of malarial toxin

The idea of malarial disease being caused by parasitesreleasing a toxin is even more venerable than that of vaso-occlusion, since it is based on a report by Golgi in 1886[59], in which he noted onset of fever and rigors at a pre-dictable short interval after the regular shower of new par-asites escape from bursting red cells. These principles were

much discussed in the first decade of the 20th century[60]. Clearly, something like this was needed to explainhow tissue not invaded by the parasite was neverthelessdamaged during falciparum malaria. Examples are sitessuch as the adult kidney and lung, where dysfunction canbe catastrophic, yet sequestration never obvious, andoften absent. The toxin idea lay fallow for many decades,not helped, in hindsight, by the underlying assumptionthat toxicity arose directly from a parasite product, in themanner of tetanus toxin.

The proposal that malarial products were not harmful inthemselves, but only through causing the infected host toharm itself through generating toxic amounts of mole-cules that, in lower concentrations, inhibit growth ofmalarial parasites, gave the toxin concept new impetus

[15,16,19]. These papers predicted that the nature of themalarial product that triggers illness could be definedthrough its ability to induce release of TNF from mamma-lian cells. A group in London did much work along theselines in the late 1980s and early 1990s [61,62], and con-cluded it was closely related to phosphotidylinositol (PI)[63]. Others extended this argument to the glycosylatedform of this molecule (GPI) [64]. The original proposalthat malarial toxin operates through inducing generationof TNF and related cytokines was greatly strengthened

when immunizing mice against GPI and then infectingthem with one of the mouse malaria parasites protectedagainst certain pathology that TNF causes on injection

[65]. Indeed, this study reports having established thatGPI appears sufficient and necessary for the induction bymalarial parasites of host pro-inflammatory responses invitro. The field has been well reviewed recently [66], withthese authors and others expressing doubts about the wis-dom of vaccinating against GPI to prevent malarial dis-ease [21,66,67]. As noted some years ago, the need forsufficient TNF to allow immune activation to proceed nor-mally during infections is plausibly why this potentiallylethal mediator has survived 300500 million years ofevolution [68]. However, despite recent reaffirmation ofthe GPI/cytokine/disease concept [69], the group that firstsuggested that GPI was the main TNF inducer in malaria

appear to have recently [25] changed their disease modelto one that eliminates a requirement for inflammatorycytokines. In view of GPI having been identified throughits capacity to induce pro-inflammatory cytokines, it

would have been remarkable to chance upon a moleculethat induces these mediators, yet mimicks their actions intheir absence. Clarification awaits a more detailed report.

Breadth, and acceptance, of the cytokine concept of

disease pathogenesis

The extensive parallels that exist between the sepsis andmalaria literature can be viewed from the perspective ofthe wide range of functionally-important inflammatory

cytokines present in the circulation in both conditions(Table 1). This strengthens the view that the two diseasesoperate through very similar mechanisms. Nevertheless, agroup working with African children has recently advo-cated that in order to understand falciparum malaria dis-ease one must return to the pre-cytokine era. Theyevidently still espouse the idea that local vaso-occlusionuniquely sets the organ pathology of this disease apartfrom others with which it is clinically confusable, in par-ticular sepsis [3-5,70]. Since the failure of treatment withcorticosteroids to ameliorate severe cerebral malaria has
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


5/32



been used as evidence against cytokine involvement [4], itwarrants recalling that MIF, known to be high in this cir-

cumstance [71], antagonizes glucocorticoids [72], andnitric oxide (noting iNOS is also high [71]) inhibits glu-cocorticoid binding to its receptor [73]. Moreover, thisdata rationalizes the failure of corticosteroid as a treat-ment.

Analogy with other diseases is still an under-exploitedtool in malaria. The original interest in TNF as a possiblemediator of both innate immunity and disease pathogen-esis in infectious disease came from analogies between theability of BCG to protect against both tumours and intra-erythrocytic protozoa [9]. When introducing the excessinflammatory cytokine concept in 1981 [15], a common

case for malaria, gram-negative bacteria and the Jarisch-Herxheimer reaction, all of which withstood the test oftime, was argued by analogy. As reviewed recently [21],the range of infectious diseases that come under the sys-temic inflammation umbrella now extends beyond bacte-rial diseases to those caused by rickettsias, protozoa otherthan malaria, and viruses. Moreover, increased circulatinglevels of TNF and functionally similar cytokines have beenmeasured in the serum very soon after onset of illness in

virtually all those infectious diseases in which it has beensought. In addition, essentially all of the signs and symp-toms involved in the clinical confusability of malaria andother causes of fever were inadvertently reproduced dur-

ing the era when rTNF was being injected in volunteers asan antitumour agent [74,75]. This includes headache,fever and rigors, nausea and vomiting, diarrhoea, ano-rexia, myalgia, thrombocytopaenia, immunosuppression,coagulopathy and central nervous system manifestations,all of which have a literature on a mechanism throughinflammatory cytokines. The rate, timing and intensity ofcytokine (pro- as well as anti-inflammatory) release will

vary in different disease states, and also between individ-uals, and provide them with somewhat distinctive clinicalpictures, but the fundamentals remain. The clinical pat-

terns generated are remarkably close, in that, at least insome populations, clinical features cannot predict a diag-nosis of malaria from other causes of fever [76].

The principle extends beyond infectious diseases. A

number of non-infectious states fit this pattern, withexcessive release of pro-inflammatory cytokines produc-ing a systemic inflammatory response. As in malaria andsepsis, metabolic acidosis [77,78], hyperlactataemia[79,80] and encephalopathy are seen in tissue injury syn-dromes such as heatstroke, trauma, and burns. Asreviewed [21], all of these conditions are ripe for an expla-nation in terms of HMGB1, liberated from the nuclei ofdamaged tissue [81], setting the scene for a broad range ofinflammatory cytokine release. Iatrogenic cytokine releasesyndromes, such as the side effects of OKT3 therapy [82],and acute graft versus host disease reaction [83,84] canalso exhibit these changes, including a reversible enceph-

alopathy. In both of these conditions the relevant pro-inflammatory cytokines are produced excessively, and

where tested (side effects of OKT3 therapy [85], and acutegraft-versus-host disease reaction [86]), prior exposure toneutralizing antibody directed against TNF prevents ill-ness.

As with much research on neutralizing antibody to TNF,this outcome does not imply that TNF is more importantthan, for example, IL-1 in this context, since anti-IL-1 anti-bodies have rarely been tried. Blocking IL-1, and indeedIL-1 and TNF simultaneously, are in their infancy, butshow promise [87,88]. Likewise, research into the disease

aspects of LT, present in falciparum malaria, and relevantto the mouse model of cerebral malaria, is relativelyignored, largely through difficulty of obtaining reagents.

Additional strong evidence for inflammatory cytokinesand falciparum malaria being functionally intertwinedcomes from studies on variation in the human genome in

Africa [89-91]. It is now accepted that falciparum malaria,historically the major fatal endemic disease in much ofthis continent, is associated with polymorphisms of thesepro-inflammatory cytokines and iNOS, which are inducedin this disease. Not surprisingly, sepsis and meningococ-cal disease have a similar literature [92,93]. Like any otherDNA trying to survive, that of humans uses trial and error

to adapt itself to its surroundings, leaving a trail of evi-dence as it does so.

In summary, illnesses arising from excessive systemic pro-duction of inflammatory cytokines include not justmalaria and sepsis, but many more infectious, and non-infectious, diseases. Insights gained by recognizing the

value of argument by analogy across this wide spectrumhave been immense, and the general concept is now sofirmly entrenched that, as noted earlier, its influence onthe evolutionary effects of infectious disease is a research

Table 1: Some changes common to systemic inflammatory

states in general, including sepsis and falciparum malaria

TNF, IL-1, iNOS and IFN-gamma, MIF, IL-10 and HO-1 raised

gamma-delta T cells raised

MRP8 (S100A8) and MRP14 (S100A9) raised

Procalcitonin raisedHMBG1 raised

ICAM, VCAM and p-selectin raised

insulin resistance

hyperlactataemia

hypoglycaemia

metabolic acidosis

hyponatraemia

coagulopathy

thrombocytopaenia

decreased red cell deformability
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


6/32



topic in its own right [26]. It would, therefore, be mostunexpected were the illness of falciparum malaria, so clin-ically confusable with other infectious diseases, andknown to generate the same inflammatory cytokines asthey do, to arise from an unrelated mechanism [5,25].

Mitochondria unable to use available oxygen, aprimary effect of inflammationInflammatory cytokines reduce ability of mitochondria to

use oxygen

One of the many actions of the cytokines responsible forsystemic inflammation is to disable oxidative phosphor-

ylation within mitochondria. This is reflected in thehyperlactataemia commonly seen in severe infectious dis-ease, and correlating with outcome. This is not to suggestthat oxygen supply and its utilization are not often limitedsimultaneously, and interact. Indeed, if toxin is replacedby its downstream consequence (the effects of pro-inflam-

matory cytokines) such interaction was proposed formalaria by Meleney over 60 years ago [94].

Both sepsis and malaria researchers have shown thatinjecting TNF, the prototype inflammatory cytokine,increased in both diseases, causes hyperlactataemia[95,96], and blood lactate levels in severe malaria haveproved to correlate closely with levels of both TNF andinterleukin-1 [97]. Nevertheless, it is fair to say that sepsisresearchers, without the tradition of a primary role forsequestration to defend, have been more receptive thanmalariologists to the systemic inflammatory explanationfor altered carbohydrate metabolism, and more readily

pursued it. Thus, they were more open than most of theirmalaria counterparts for the insights of the early 1990s,

when newer techniques demonstrated that oxygen ten-sion in tissues was actually increased (not decreased, asthe literature predicted) in septic rats [98], patients [99],and pigs [100]. This implied an inability of mitochondriato utilize oxygen, forcing glycolysis to compensate, as bestit can, for the energy deficit. The next insights came fromgroups who developed the cytokine-induced mitochon-drial dysfunction model of disease [101-103] and, thus,provided an inflammation-based explanation for a shut-down of aerobic glycolysis, a consequent increased rate ofglycolysis, and thus lactate production, metabolic acidosis

and cellular energy depletion. In any disease with highlevels of inflammatory cytokines this mimics poor oxygensupply. An important difference, however, is that theeffects of the nitric oxide through which mitochondrialshutdown largely operates are reversible [104,105],

whereas frank hypoxia, through vaso-occlusion, as evi-denced by the stroke literature, is less so. Thus, the former,not the latter, is consistent with the marked reversibility ofmetabolic comas [106], a term advocated [21,107] toinclude human malaria.

This approach to understanding energy balance in sepsishas been followed successfully on a number of cell types,including hindlimb skeletal myocytes, gut wall cells, andhepatocytes. The wide range of tissues in which these con-cepts have been demonstrated adds to the arguments on

systemic origins of lactate in sepsis and malaria. For exam-ple, inflammatory cytokines have been shown to causecontractile dysfunction [108,109] and also energy deple-tion [110,111] through effects, often mediated throughinduced nitric oxide [109,112], on cardiac muscle. Like-

wise, in diaphragmatic skeletal muscle there is evidence ofcytokine-induced nitric oxide [113,114] and oxygen-derived free radicals [115] combining to form peroxyni-trite [116,117] and this causing dysfunction of mitochon-dria in myocytes, leading to energy depletion and thusmuscular contractile failure. The outcome here is toreduce the patient's ability to counter acidosis by blowingoff CO2.

An additional pathway through which the inflammatorycytokines may reduce oxygen consumption is throughperoxynitrate (OONO-), a product of NO (from iNOSinduced by these cytokines) and superoxide, overactivat-ing poly(ADP ribose) polymerase-1 (PARP-1) [118]. Thiscan deplete cellular stores of NAD+, and efforts to resyn-thesise it can deplete ATP as well (reviewed in reference[119]). Moreover, NAD+ is essential for glycolysis, so itsdepletion can be expected to impair glycolytic input intomitochondria. These concepts were reviewed in depth ina malaria context a few years ago [107].

Mitochondria starved of oxygen, a secondaryeffect of inflammation

This section summarizes the ways in which inflammatorycytokines indirectly limit the supply of oxygen to cells,and thus further reduce the capacity of their mitochondriato generate ATP through oxidative phosphorylation.

There are good arguments from the basic literature thatthey may do so through directing sequestration towardsorgans that are particularly oxygen-sensitive. Being a newconcept in the malarial world, this literature is examinedhere in some detail. In addition, anaemia, cardiac insuffi-ciency, or insufficient circulating volume (see below, andFigure 3) can all be driven by inflammatory cytokines.

Again, infectious disease in general is outlined beforefocussing on the particular case of falciparum malaria.

Inflammatory cytokines cause blood elements to adhere to

endothelium

It is well accepted that upregulation by inflammatorycytokines of adhesion sites on endothelial cells invitessusceptible circulating blood elements to attach to theinner wall of blood vessels. In many diseases, includingmalaria, this includes activated leukocytes and platelets,both of which play important roles in promoting proco-
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


7/32



agulant activity. For example, in malaria this activity isseen on circulating monocytes [120] and placental macro-phages [121], and the thrombin so formed enhancesadhesion by increasing expression of CD36 on plateletsurfaces [122].

Platelets and leukocytes

Platelets and leukocytes have been reported to adhere toendothelium in viral [123-125], bacterial [126,127] andprotozoal infections, including the cerebral vasculature inpaediatric falciparum malaria [71,128]. In particular,aggregated monocytes are a striking feature in malarialplacentas [129], and a less dramatic finding in cerebral

vessels [130], where they and the nearby endothelial cellsstain strongly for iNOS [71]. These adhering elements canset up local foci of inflammation, generating more inflam-matory cytokines (eg TNF in placentas [131]), includinginflammatory cascades initiated by HMGB1 released from

the adhering activated platelets [132]. Since HMGB1increases are associated with severity of falciparummalaria [57], this could account for the potentiatingeffects of platelets reported in an in vitro model ofendothelial activation byP. falciparum [133]. Along withthe effects of systemic inflammation, these local inflam-matory foci contribute to potentially fatal pathology,including loss of endothelial integrity, in infectious dis-eases in which circulating inflammatory cytokine levelsare sufficiently increased. Notably, the range of complica-tions seen in falciparum malaria, including coma, cantherefore develop in sepsis [134], influenza [135], andsometimes vivax malaria [136,137], which show evidence

of a high systemic inflammatory response, but have nopossible involvement of sequestering parasites.

Parasitized red cells, an additional circulating adherent element in

falciparum malaria

Although falciparum malaria shares with conditions suchas severe bacterial disease much evidence for endothelialactivation (eg adherence molecules [138,139] and circu-lating endothelial microparticles [140,141]) and its con-sequences (eg platelet and leukocyte adherence, above), itis distinguished from them by the presence of anotherobvious adherent object erythrocytes containing matureparasites. From the biology of the erythrocytic phase ofP.

falciparum, vascular sequestration (endothelial adher-ence) of parasitized red cells, somewhere in the circula-tion, is inevitable for roughly the last half of the 48 hrerythrocytic cycle. Thus mature erythrocytic forms of theparasite are rarely seen in peripheral blood smears. Thisadherence was first noted in autopsy samples in the 19thcentury [142], and fuelled the widely (but not universally;see below) held view that much of the illness and pathol-ogy of this disease needed little explanation other thanthat offered through consequential impairment of micro-

vascular flow. Thus, through the decades a common

thread in proposals to explain falciparum malaria diseasehas been a primary role for tissue hypoxia caused by vaso-occlusion by parasitized red cells. The presence of coma,hyperlactataemia, hypoglycaemia, and metabolic acido-sis, all three consistent with a patient being forced to rely

on anaerobic glycolysis for energy production, haveencouraged this viewpoint. As summarized earlier,cytokine-induced cytopathic hypoxia can also explainthese phenomena.

Microparticles in systemic inflammatory diseases

Microparticles are an intriguing component of the inflam-matory system. First described in 1967 [143], they are aheterogeneous group of small membrane-coated vesiclesreleased from many types of cells upon their activation orapoptosis, but differ from apoptotic bodies. They retain atleast some functions of their cell of origin, which caninclude platelets, endothelial cells, and various leuko-

cytes. Triggers for their release include TNF [144], andthey are increased in the circulation in systemic inflamma-tory states such as sepsis and trauma [145,146], as well asinducing inflammatory cytokine release themselves [147].

Along with cytokine increases, endothelial activation andthe activation and adhesion of platelets and leukocytes,microparticles provide further evidence for a commonpathophysiology of sepsis, trauma, and malaria[140,148,149]. While microparticles enhance the generalinflammatory activity in these circumstances, and theirexploration within the context of malaria is novel, evi-dence that they are likely to prove to be a key to control-ling malarial disease, any more than other inflammatory

conditions in which they act as markers of severity, is sofar lacking.

The interaction between cytokines and sequestration

Arguments against a primary role for sequestration in falciparum

malaria illness

As noted earlier, sequestration is common in certain tis-sues of fatal cases of falciparum malaria. Nevertheless, aprimary (ie prior to cytokine increase) harmful vaso-occlusive role for sequestered red cells containing para-sites does not, in our view, withstand close scrutiny. As asimple practical example, it requires the pathophysiologyin patients equally ill from uncomplicated falciparum

malaria and vivax malaria to be quite different. Parallelsbetween vivax malaria and the outcome of injecting TNFinto human volunteers [75,150], and increased levels ofthis and functionally-related cytokines in vivax patientsera [151] plus the non-sequestering nature of the parasi-tized red cells, are consistent with cytokines being suffi-cient to cause the illness of vivax malaria. This includesthe occasional, but well documented, coma (see below).Logically, therefore, these cytokines are sufficient toexplain uncomplicated falciparum malaria, a conditionnotoriously difficult to separate clinically from vivax
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


8/32



malaria [76]. This is consistent with occlusive parasitizedred cells becoming an essential part of pathogenesis only

when falciparum disease becomes more severe, and para-sitized erythrocytes start to favour physiologically sensi-tive areas such as the cerebral vasculature. Alternatively,

the total number of parasites, sequestered or otherwise,may, through the cytokine concentrations they generate,be what counts.

The erythrocytic life cycle dictates that sequestration isinevitably present before the onset of illness or in malariatolerant individuals, yet no information seems availableon it taking any particular anatomical pattern in thesepeople. Newer technology involving expression of luci-ferase by a defined stage of the life cycle, to date appliedonly to a mouse model (where it revealed intriguingresults that warrant reading closely [152]), would beimpractical in man. However, an indirect appreciation of

the avidity, if not the location, ofP. falciparum sequestra-tion in asymptomatic, malaria-tolerant individuals (usu-ally not even fever, yet parasite densities overlappingthose seen in severe illness) comes from work done inMali [153], in which blood smears were made three timea day, usually at 6 hr intervals, for 1213 days. Some 3000blood smears were examined. Fluctuations in parasitedensity between consecutive smears often proved to bemassive and abrupt, and the authors concluded that suchchange could come only from mirror-image changes insequestration rate, not through parasite multiplication.

Transitory disappearance of parasites from the peripheralblood occurred at least once in all infected individuals (63

of 79 subjects). Such slide negativity was of short dura-tion, unrelated to the cycle of trophozoite maturation,and attributed to dramatic increases in sequestration ofexisting parasites. To our knowledge this rapid oscillationbetween circulation and sequestration is not reported dur-ing untreated malarial illness. On this evidence, avidity ofsequestration is lower in malaria-tolerant individualsthan in patients, suggesting different controlling influ-ences, with sequestration in the former group, but not inthe later, being independent of inflammatory cytokines.

At autopsy, brain is often a favoured site of sequestration,but whether this preference occurs before coma onset, or

develops while coma is progressing, has not been deter-mined. Marked brain sequestration at death has not beena universal observation, and, as summarized by Maegraith[18], reports have accumulated since the 1920s on a mis-match, at autopsy, between cerebral sequestration andcoma. Such evidence is still being presented [71], as dis-cussed below. Impaired consciousness can occur in arange of systemic inflammatory states, being present incertain viral and bacterial diseases as well as malaria. Thusregarding parasite sequestration as a necessary mecha-nism for falciparum malarial disease ignores its close clin-

ical and pathological similarity, in terms of metabolicchanges and organs affected, to other diseases that alsocan cause impaired consciousness, but lack parasitized redcells. These conditions are now accepted to be systemicinflammatory states [76]. As summarized below, the

broader literature is consistent with our novel proposal(below) that, in addition to the above post-sequestrationactivity of inflammatory cytokines, these mediators willhave earlier determined, during severe illness, where mostsequestration occurs.

For some time most adherents of the traditional seques-tration-based view of falciparum malaria disease haveaccepted what they regard as secondary roles for inflam-matory cytokines in falciparum malaria disease [154]. Bythese tenets, sequestration at sensitive sites, such as braincapillaries, leads to higher local concentrations of thesecytokines near sequestered parasitized red cells, since

these are the source of the parasite material that triggerstheir release, and endothelial cells and leukocytes arecommonly their source. The importance of the site ofsequestration to disease severity is thereby amplified,involving both vaso-occlusion and secondary cytokineeffects, such as increased blood-brain barrier permeability[155]. This as a plausible aspect of the relationshipbetween sequestration and inflammatory cytokines in fal-ciparum malaria, but the concept outlined below is pro-posed here to be their major interaction in falciparummalaria.

Influence of inflammatory cytokines on organ distribution of

sequestrationBy far the bulk of the literature arguing for occlusion-induced pathology in falciparum malaria concerns itsdocumentation in the brain [156] and placenta [157] insick individuals, and it is timely to consider why thesesites are favoured. Equally, why are these the main sites

where monocyte accumulations occur? The followingexplanation seems highly plausible, and is testable. Inbrief, it suggests that sequestration favours the brain whencirculating concentrations of TNF are high (ie when thepatient is ill), but not before onset of illness, or in malariatolerant individuals. The absence of mature forms onblood smears from these two groups implies that the

capacity of parasitized red cells to sequester is quiterobust, but sequestration is not focussed at harmful sites

without raised inflammatory cytokines.

For over a decade there has been substantial evidence thatinflammatory cytokines (TNF the most studied) increaseexpression on endothelial cells of the molecules to whichparasitized erythrocytes adhere [158-160]. Being drivenby cytokines whose detection and concentrations corre-late with degree of illness, this increase can be expected tooperate only in moderate to severe illness, not early in
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


9/32



infection, when TNF levels are undetectable, or in tolerantindividuals, who have been argued to be refractory tomalaria-induced TNF [161]. Blood smears confirm thatsequestration occurs in these individuals. In the next par-agraphs it is argued that during severe falciparum illnessthese cytokines (provided they have not killed the patientbeforehand) could concentrate most sequestration to thesites familiar at autopsy.

TNF and interleukin-1 increase tissue factor expression onendothelial cells and mononuclear cells [162], therebyinitiating pathways that generate thrombin (reviewedrecently [163]), a molecule with many roles at the crossroads of inflammation and coagulation. When bound tothrombomodulin, a thrombin receptor on the endothe-lial cell surface, thrombin activates protein C, which candegrade Factor VIIIa and Factor Va, essential cofactors in

The proposed influence of differences in thrombomodulin levels on cytokine-induced expression of adhesion molecules onendothelial cells, and monocyte attraction, in different organsFigure 1The proposed influence of differences in thrombomodulin levels on cytokine-induced expression of adhesion molecules onendothelial cells, and monocyte attraction, in different organs. (a) tissues with low endothelial thrombomodulin levels (b) tis-sues with high levels.
http://-/?-http://-/?-http://-/?-http://-/?-


10/32



the activation of Factor X and prothrombin respectively[164]. These feedbacks play a central role in keeping coag-ulation in homeostasis.

It follows, therefore, that tissues in which thrombomodu-lin density on endothelial cell surfaces is lowest (brainleast indeed reported undetectable in an earlier study[165] placenta next least, and other organs more [166]),

will have more unbound thrombin left available for itsother functions on activated endothelium. These includea ten-fold upregulation of adhesion molecules such as E-selectin by TNF [167], CD36 [122], intercellular adhesionmolecule-1 (ICAM-1 CD54) and vascular cell adhesionmolecule-1 (VCAM-1 CD106) [168] and, with implica-tions for the observed accumulation of monocytes, mono-cyte chemotactic protein-1 (MCP-1) [169]. Moreover,

thrombin thrombomodulin complex formation will below in tissues where endothelial thrombomodulin is low.

Therefore protein C activation will be correspondinglylow, and the negative feedback that controls TNF-inducedthrombin formation correspondingly weak (Figure 1a),further enhancing concentrations of the above adhesionmolecules. Levels of a range of inflammatory cytokines,including TNF, are high in supernatants of villous leuko-cytes from malarial placentas [170], so these principlesshould apply to the monocyte accumulations and heavysequestration in this organ also.

Thrombomodulin also sequesters HMGB1, making it lessavailable to activate RAGE (the receptor for advanced gly-cation endproducts, shared by this cytokine [171]), so it

cannot express its full inflammatory potential [172], andthus generate a further wave of cytokines such as TNF.Hence a given concentration of HMGB1, a cytokineincreased in serum in sepsis [53] and falciparum malaria[57] in proportion to degree of illness, can be predicted toexert more pro-inflammatory influence in brain and pla-cental vessels, where more of it is functionally availablebecause less of it sequesters on thrombomodulin. Unfor-tunately the intestinal blood vessels, another favoured sitefor sequestration in falciparum malaria, were notincluded in either the CD36 [122] or the thrombomodu-lin study [166]. The reverse of the arguments for brainserve to rationalise why sequestration is rare or absent in

heart and skeletal muscle, tissues at the other (high) endof the thrombomodulin spectrum [166], and therefore

with least free thrombin left available to upregulatesequestration sites during TNF-induced illness (Figure1b). For these reasons harmful sequestration is bestregarded as a consequence of increased inflammatorycytokine generation, as well as a potential way to focusrelease of these mediators at sequestration sites. In short,differential endothelial activation induced by high levelsof circulating inflammatory cytokines could shift theemphasis of sequestration to potentially harmful loca-

Immunohistochemical staining of the gut wall of malaria patients to detect iNOSFigure 2Immunohistochemical staining of the gut wall of malaria patients to detect iNOS. Techniques (DAB, haematoxylin), materialsand controls as in reference 71. Cases (a) MP6 and (b) MP21 (see Table 1 of ref. 71) are shown. Unpublished data.

(a) (b)
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


11/32



tions where, through schizogony, it could then initiate thebulk of the next wave of cytokine release. As previouslynoted for neuron function in malaria [173], nitric oxidegenerated by iNOS induced by these cytokines could alsoplausibly explain the small intestine intussusception seen

in most children dying with malarial coma in Malawi[174]. Nitric oxide has an essential role in an experimentalmodel of this pathology [175], and iNOS is stronglyinduced in the small vessels of the jejunum (Figure 2) inthis patient series.

Anaemia

As recently reviewed [176], critical illness associated withan inflammatory response invariably causes multifacto-rial anaemia. It has often been noted that anaemia couldcontribute to poor oxygenation of tissues in malaria [177]and there is general acceptance that it can be severeenough to reduce supply of oxygen to mitochondria to

dangerously low levels. Thus it can be a major componentof malarial pathology. Obviously a high parasite loadindicates imminent widespread lysis, but anaemia doesnot correlate with parasitaemia, and sometimes is extreme

when very few parasites are present.

Poor red cell deformability

Erythrocytes have a limited life, determined by how longthey can remain flexible enough to squeeze through fen-estrations in specialised vessels in the red pulp of thespleen [178]. A red cell that cannot pass this test is phago-cytosed by adjacent macrophages, and lost. In health thisloss is balanced by erythropoiesis, and haematocrit

remains normal. Should red cells develop a prematurepoor deformability they are removed from the circulationcorrespondingly earlier.

Like other cells, erythrocytes stay intact by constantlyextruding Na+ in exchange for K+ through an energy-dependant "pump" in their cell membrane that wasdefined by the ability of certain digitalis gylcosides toblock it. This Na+/K+ pump fails, and intracellular Na+

accumulates in (non-parasitized as well as parasitized) redcells during human [179] or monkey [180] malaria. Thesechanges in ionic content of red cells have been observedin a sepsis model [181]. In another sepsis model [182],

erythrocyte deformability could be shown to be caused byNO, an inhibitor of this membrane pump [183]. Sinceinhibition of the Na+/K+ pump in vitro correlates with areduced red cell deformability plus a parallel decrease inred cell filterability [184], any influence, such as NO[183,185], that inhibits this pump could potentially causepoor red cell deformability. Cytokine-induced iNOS pro-

vides a demonstrable [71] way for these changes to occurin severe malaria.

Originally recorded in uraemic patients, poor red celldeformability was observed in a small pilot study ofmalaria patients in 1985 [186]. Soon after it was recog-nized in sepsis [187,188], and subsequently studied in fal-ciparum malaria with a view to understanding both

circulatory obstruction [189] and anaemia [190]. There isgood evidence that, when measured on admission, asevere reduction in red cell deformability is a strong pre-dictor of malarial mortality [189], but whether this iscause and effect, or the two phenomena are simply inevi-table co-travellers in a strong pro-inflammatory milieu, isunclear. It seems clear that poor red cell deformability(which affects parasitized and unparasitized red cellsequally) and dyserythropoiesis can lead to severe anaemiain various diseases, particularly in chronic infections suchas malaria. Its presence in vivax malaria [191] implies thatits role in vaso-occlusion is less important.

Clearly, it would be useful if a shortened lifespan of redcells through their premature loss of membrane flexibility

were compensated for by a faster rate of erythropoiesis.Unfortunately, the inflammatory cytokines that shortenthe lifespan of the red cells also slow down their replace-ment, as outlined in the next section. Their combinedeffect on reducing haematocrit can be expected to berapid.

Dyserythropoiesis

Because the parasite inhabits erythrocytes, which mustburst if the parasite is to propagate, the obvious initialconclusion was that this source of red cell loss was central

to the fall in haematocrit seen in this disease. As reviewednearly 60 years ago [192], this fall was soon realised to beout of proportion to the number of red cells parasitized,so other factors were realised to contribute. Phagocytosisof unparasitized red cells was also recorded decades ago inmonkey [193] and human [192] malaria, and for many

years was regarded as sufficient explanation for this dis-crepancy. Others had been investigating dyserythropoiesisin the bone marrow of patients with falciparum malaria[194,195] and stressed its contribution to malarial anae-mia. A group in Oxford [196], seeking an explanation forthis dyserythropoiesis through an electron microscopystudy of bone marrow, observed sequestration of parasit-

ized red cells and argued that this caused the bone marrowdysfunction in falciparum malaria by restricting bloodflow and thus inducing hypoxic changes. This idea provedinadequate, however, when this same group subsequentlyreported dyserythropoiesis and erythrophagocytosis in

vivax malaria, in which parasitized red cells do notsequester [197].

Twenty-five years ago our group proposed that TNF mightcause the bone marrow depression seen in malaria [15].Subsequently an undefined product in macrophage super-
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


12/32



natants [198], later identified as TNF [199], was found toinhibit the growth and differentiation of erythroid pro-genitor cells. When rTNF became available (but before ithad become technically possible to assay for this cytokinein human serum) the dyserythropoiesis and eryth-

rophagocytosis seen in terminal Plasmodium vinckei-infected mice were reproduced when a single injection ofrTNF was given early in the course of the infection [200].Phagocytosis of erythroblasts in bone marrow, a phenom-enon also reported by Wickramasinghe et al. [196,197] inhuman malaria, was commonly observed [200].Decreased erythropoiesis was subsequently reported inmice receiving continuous TNF infusions via implantedosmotic pumps, and increased erythropoiesis in malarialmice after injecting neutralizing antibody directed againstmurine TNF [201]. TNF-induced dyserythropoiesis hassince been confirmed in rats [202], and mice expressinghigh levels of human TNF become markedly anaemic dur-

ing malaria infections [203], even though parasite num-bers, and therefore red cell loss post-schizogony, areconsiderably reduced.

The past decade has seen an expansion of this line ofenquiry into human malaria, and also the number ofcytokines, both pro-inflammatory and anti-inflammatory[204,205] in absolute amounts and ratios [206,207], thathave been investigated in this context. It has beenextended to include other pro-inflammatory cytokines,such as IL-12 [208] and FasL [40], and the role of the per-sistence of production of such cytokines in the anaemia offalciparum malaria infection has recently been examined

[209]. Suppression of prostaglandin E2 during malariainfection has also been shown to have an important influ-ence on these events [210].

A decade ago the mechanism of TNF-induced damage tohuman bone marrow cells was argued to be nitric oxidegenerated by iNOS induced by TNF [211]. More recentlyattention has focussed on another cytokine, MIF, down-stream of TNF but also induced by agents other than TNF,as a cause of malarial dyserythropoiesis. Martiney and co-

workers [212] found that MIF was enhanced in a mousemalaria model, and that rMIF inhibited the formation oferythroid (BFU-E), multipotential (CFU-GEMM), and

granulocyte-macrophage (CFU-GM) progenitor-derivedcolonies in vitro. Subsequently, MIF proved to be stronglydetectable by immunohistochemistry in systemic, but notcerebral, vascular smooth muscle of fatal African paediat-ric sepsis and falciparum malaria [71]. It has very recentlybeen found that sub-inhibitory concentrations of MIFsynergise profoundly with TNF and interferon-gamma ininhibiting mouse erythroid precursor colonies [213].

These authors provide other data that greatly strengthensthe case for a major role for MIF in malarial dyserythro-pioesis. This work provides a timely warning against the

reductionist approach to understanding the actions ofcytokines in disease, which does not reflect the in vivo real-ity of a considerable number of these mediators beingpresent simultaneously. Thus the slow replacement rate ofred cells in malaria through the influence of inflammatory

cytokines is now a well-established aspect of malarial dis-ease pathogenesis. In summary, cytokines induced bymalaria products are a major determinant of haemo-globin deficiency, and thus the rate at which oxygenreaches mitochondria in malaria.

Infection-induced dyserythropoiesis is not restricted tomalaria. The first awareness of it in other infectious dis-eases appears to have been its description in HIV patients,plausibly as a consequence of opportunistic infections[214]. It has subsequently been observed in acute viralhepatitis B [215], simian [216] and human [217] parvovi-rus B19, visceral leishmaniasis [218] and dengue [219], all

conditions that are associated with increased levels of cir-culating TNF, and doubtlessly its regulators, such as MIF.

As noted above, the effect on red cells of the combinationof a lower rate of production and accelerated destructioncan be expected to lead to severe anaemia. The literatureon both these influences on red cells underline how

widely the consequences of excessive inflammatorycytokines impinge on disease pathogenesis, and empha-sise the conceptual limitations imposed by regarding fal-ciparum malaria as somehow outside the sphere of thesehost-origin mediators [5,25].

Cardiac insufficiency

Cytokine-induced myocardial depression frequentlyaccompanies severe sepsis (see [220]). Whereas it was pre-

viously considered a pre-terminal event, it is now clearthat cardiac dysfunction, as evidenced by biventricular dil-atation and reduced ejection fraction, is present in mostpatients with severe sepsis. It has been known for sometime to be caused by soluble factor(s) released by macro-phages exposed to endotoxin [221]. Once cloning ofcytokines occurred its activity was attributed to IL-1, thenalso to TNF [222], then the two synergistically [223], andfinally to IL-6 [224], a macrophage product induced byboth of these mediators. A literature exists on these effectsbeing minimised by blocking MIF, which reduces the

feedback inhibition of TNF production by glucocorticoids[225,226]. As discussed in above, excess inflammatorycytokines have been shown to cause cardiomyocyte mito-chondrial dysfunction. Since TNF [38], IL-1 [38], IL-6 [39]and MIF [71] are all highly expressed in falciparummalaria, it can be expected that these cardiac-depressingactivities would be acting in this disease as well as in sep-sis. Evidence of this, in terms of circulating cardiac pro-teins, has accumulated in the past few years [227,228],although its clinical impact is yet to be evaluated. It maybe present, but its potential clinical impact is over-ridden
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


13/32



by the effects of hypovolaemic shock, as summarized inthe next section.

Poor circulating volume

Insufficient intravascular volume is ultimately of concern

in disease because, through poor perfusion, it leads topoor oxygen supply where it matters, the microcirculationthat feeds the mitochondria within the cells that form thetissues the capillaries pass through. The major therapeuticoption is volume resuscitation. As recently reviewed[229], it is under the control of a number of autoregula-tory mechanisms, and these are known to be disrupted insepsis. As well as the effects of changed red cell deforma-bility, and adherence of platelets and leukocytes, as dis-cussed above, variation in iNOS induction, leading tomore or less nitric oxide, essential for local degrees of the

vasodilation that perfusion depends on, are major con-trolling factors.

Using a range of indicators, workers in Kenya have con-firmed the older observation [230] that shock is not rarein severe falciparum malaria [231], and that the haemody-namic changes in children with severe malarial anaemiacomplicated by the respiratory distress were more charac-teristic of hypovolaemia than of biventricular failure[232]. They have also demonstrated that while adminis-tering albumin did not improve acidosis, it did reducemortality [233]. This conceptual approach has been stren-uously questioned by others who detected only a mild fallin total body water volume and extracellular water vol-ume [234], as well as the relative rarity of severe hypoten-

sion in falciparum malaria compared to the shock that canaccompany trauma or sepsis [4]. However, local effectsmay be much more important in falciparum malaria thanin sepsis for example these could in part arise from

vasodilation being much more uneven in malaria thansepsis because of patchy local foci of post-schizogonymalaria toxin release from sequestered parasites, and localgeneration thus of the inflammatory cytokines that induceendothelial iNOS [71]. It is recognized that during treat-ment one would need to be cautious of fluid overload ifthe patient displays evidence of cerebral oedema [235] orcardiac insufficiency [236].

The detailed arguments on both sides of this debate arebeyond the scope of this review, except to note that theyare currently a major fault line between those who pro-pose that malaria has a fundamentally similar pathophys-iology to other acute systemic infections [21,233,237] andthose who see it as unique [238]. This issue cannot beresolved until recognition is given to the need to researchthe pathophysiology of malaria and other systemic infec-tious states in parallel rather than, as at present, in isola-tion.

From this and the previous section, it is not hard to visu-alise the combined harmful effect on the patient whensystemic inflammation reduces oxygen supply to theircells also makes these cells worse at using it. As shown inFigure 3, the initiating pathophysiological lesion is the

onset of the systemic inflammatory response, and it is dif-ficult, from the evidence, to envisage sequestering parasit-ized red cells, per se, initiating malarial disease before it isfocussed to sensitive organs by systemic release of inflam-matory cytokines. Sequestering parasitized red cells maythen, in part through locally released cytokines, exacer-bate the illness if the patient survives long enough.

Practical consequences of these changesHyperlactataemia in malaria and other infectious diseases

Hyperlactataemia, a recognized marker of falciparummalaria severity, is at the centre of controversies relevantto the theme of this review. Its discussion requires some

basic biochemical background. The lactate anion hascomplex roles in biology. Hyperlactataemia may be asso-ciated with acidosis, a normal pH, or alkalosis [239], andcan occur in viral and rickettsial diseases [240], as well as(see below) sepsis and malaria. In synopsis, most lactateis generated during glycolysis, which essentially consistsof oxidising glucose, a six-carbon structure, into two three-carbon molecules of pyruvate. This is reduced to lactatethrough the action of pyruvate dehydrogenase, a reactionthat avoids pyruvate accumulating, and supplies NAD+ tokeep glycolysis going. Thus lactate can be formed as abyproduct of glycolysis, which can occur in all metaboli-cally active tissues and supplies ATP, albeit in small

amounts, independently of the presence of oxygen. Everymole of glucose metabolised by anaerobic glycolysis tocarbon dioxide and water yields 4 moles of ATP, whereasoxidative phosphorylation within mitochondria yields 32moles of ATP. When oxygen usage falls (whether throughpoor supply or poor utilisation) ATP generation falls, andglycolysis is accelerated to compensate, as much as possi-ble, for this energy loss. A consequence is oversupply ofthe byproduct, lactate, but from a disease perspective thisis a side issue compared to insufficient ATP generation,even though the two may correlate well. Enhanced glyco-lysis under aerobic conditions can also increase lactateproduction. The metabolic acidosis secondary to this fail-

ure of mitochondrial energy production, which high lac-tate often accompanies, is a consequence of this energyfailure, and inevitably accompanies it in severe inflamma-tory illnesses, including malaria and sepsis.

The body's supplies of glucose, including stores of its pol-ymer, glycogen, are not unlimited, so when glycolosis isenhanced for any period it sooner or later runs out of fuel.Gluconeogenesis fills the breech as much as possible, butit soon fails because substrate supplies are limiting [241].

These events are reflected in the hypoglycaemia that has
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


14/32



often been reported in severe malaria [242] and sepsis[243,244]. When seen in this context hypoglycaemia inthese diseases is no longer a primary cause of harm, suchas when driven by hyperinsulinaemia, but an inevitableconsequence of over-exuberant, typically anaerobic, glyc-

olysis.

Origin of the high circulating levels of lactate

Poor oxygen delivery to mitochondria

In the 19th century an association had been notedbetween hypoxia and lactate accumulation in tissues, anda progression of logic through the physical exercise litera-ture [245] led to lactate levels being seen not only as amarker for poor oxygen delivery in disease states, but alsoa consequence of it, and the cause of the acidosis. Forsome time hyperlactataemia has been regarded as a func-tionally relevant marker for a poor prognosis in both sep-sis [246] and malaria [247]. It is now well accepted that

hyperlactataemia correlates with a poor prognosis in pae-diatric falciparum malaria in Africa [97,177,248].

Although the sepsis world now discusses several originsfor the lactate increase, including inflammation-inducedmitochondrial dysfunction [103], in falciparum malaria itis still generally attributed to a reduced oxygen supply,mostly through microvascular occlusion by sequesteredparasitized erythrocytes [70]. Unfortunately, the tradi-tional conceptual approach (in which not only are acidproperties attributed to the lactate anion but lactate andlactic acid are equated and used interchangeably) is dyinghard in the malaria world.

Recent publications [4,238] promote the traditional viewby arguing that lactate/pyruvate ratios are higher inmalaria than in sepsis (but see [249]), and can thereforebe explained only by hypoxia through vaso-occlusion[4,70]. However, it seems difficult to envisage a mecha-nism whereby insufficient oxygen reaching tissue mito-chondria would generate higher lactate/pyruvate ratiosthan would its poor utilisation once there. Indeed aninadvertent positive control of mitochondrial dysfunc-tion, seen as the side effects of treatment of HIV patients

with nucleoside reverse transcriptase inhibitors that causemitochondrial toxicity [250], can generate lactate/pyru-

vate ratios up to double that recently reported for severe

malaria [4,70], and as high as any value reported for theseverest adult cases in Thailand [251].

In order to test the possibility that sequestration is essen-tial for these changes in skeletal muscle sections, tissuespreviously stained for other purposes [71,252] have beenrecently re-examined for adhering parasitized red cells.Since skeletal muscle is a tissue with one of the highestrates of oxygen consumption [253], it is predictably alarge generator of lactate when anaerobic respirationdominates, whether triggered by oxygen insufficiency

from vascular occlusion by sequestering parasitized eryth-rocytes or its adequate presence, but under-utilisation,through mitochondrial dysfunction. To address the vaso-occlusion possibility, 13 tissue sections from chest walland/or diaphragm from each of 27 previously described

[71] fatal malaria cases of African paediatric malaria wereexamined blind by three observers. Negligible sequesteredparasites were observed in 24 of 27 cases (unpublisheddata), including 15 displaying a 3+ or more sequestrationscore in cerebral capillaries. In the light of the high throm-bomodulin levels on endothelial cells in vessels in skeletalmuscle [166], and its implications for expression of adhe-sion molecules (Figure 1b), this is not surprising. Unfor-tunately, lactate assays were not performed on this patientseries. Nevertheless the high incidence of hyperlactatae-mia in fatal cases from this same population [254] impliesthat the incidence of skeletal muscle sequestration wouldneed to be considerably higher than observed for this

hyperlactataemia to have its origins in impeded oxygendelivery to this main glucose-consuming organ.

Poor oxygen usage by mitochondria

The concept termed cytopathic hypoxia [103], is now con-sidered to be a major contributor to the pathogenesis ofsepsis. This is consistent with the evidence from animalmodels that neutralizing TNF prevents [255] or reverses[256] metabolic acidosis in experimental sepsis, althoughthe indirect action of TNF on oxygen delivery, through itseffects on endothelial activation, and thus platelet andleukocyte adhesion, can also be expected to contribute invivo. In addition, inhibiting TNF successfully treated the

metabolic acidosis of sepsis in a double blind trial in pre-mature infants [257], and immunizing mice against GPI,a malarial toxin selected for its capacity to induce TNFproduction, inhibited metabolic acidosis in a mousemalaria model [65]. These outcomes are equally likely toapply in any disease in which levels of the pro-inflamma-tory cytokines, including TNF, are raised, and metabolicacidosis occurs. Studies on muscle ATP depletion duringsevere sepsis in patient material [258] and experimentalanimals [259] have provided data consistent with thesearguments. It would be most instructive to repeat thesesame experiments with muscle biopsies from malariacases.

Since the inflammatory cytokines that cause mitochon-drial shutdown are prominent in both sepsis and malaria[38,39] it can be inferred that this organelle dysfunction isan equally plausible cause of reduced ATP synthesis andincreased lactate accumulation in both diseases, corre-spondingly diminishing the need to invoke the argumentbased on parasite-induced vascular occlusion to explainthese changes in malaria. Moreover, the mitochondrialultrastructural damage that correlates with lowered oxida-tive phosphorylation in a sepsis model [260] parallels that
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


15/32



described by Maegraith in monkey malaria [261]. Inrecent decades few malaria researchers have remarked onthe demonstration, by Maegraith's group, decades ago, ofthe now topical inhibition of mitochondrial function inmalaria [262,263]. In our opinion these were landmarkobservations that paved the way for a role for severe sys-temic inflammation in causing the metabolic acidosis andcontributing to the raised lactate levels seen in severemalarial disease. Indeed, in 1954 this group discussed the

concept of functional hypoxia in malaria arising frominterference with oxygen acceptance by tissue cells [264].

Hypoxia and TNF both induce HIF-1

Hypoxia-inducing factor-1 (HIF-1) is a transcription fac-tor that regulates activation of several genes responsive tolow oxygen, including erythropoietin, vascular endothe-lial growth factor, glycolytic enzymes, and glucose trans-porters. These pathways need to be switched on wheneveraerobic respiration is reduced, since they are essential forthe body to generate as much ATP as it can. As evidencethat both poor oxygen delivery and its poor utilizationthrough the inhibitory effect of inflammatory mediators

on mitochondria produce the same functional and, there-fore, clinical end results, it should be noted that TNF [265]has, even in normoxic cells, the same HIF-1-inducing abil-ity as has hypoxia [266]. This is predictable from the abil-ity of TNF to shut down mitochondria (eg via thereversible effect of nitric oxide on cytochrome c [104]),the oxygen sensor that regulates HIF [267]. It provides aplausible explanation for the accelerated rate of aerobicglycolysis sometimes reported in sepsis [268], which canincrease pyruvate and lactate levels in the absence ofhypoxia [269].

Should cytokine-accelerated glycolysis occur under aero-bic conditions, any resultant hyperlactataemia cannot beexpected to be associated with acidosis, since protons aregenerated by ATP hydrolysis in mitochondria, and pHremains constant. The desirability of enhancing HIF-1 in

systemic inflammatory states, in which category the pres-ence of excess pro-inflammatory cytokine productionplaces malaria, has recently been reviewed [270]. MIF, apro-inflammatory mediator shown to be upregulated dra-matically in a number of tissues in both severe falciparummalaria and sepsis [271], also accelerates glycolysis [272],so can be expected to contribute to the hypoglycaemiaand hyperlactataemia of both diseases. As with HIF-1,both hypoxia [273] and TNF [42] upregulate MIF. Sinceanti-MIF antibody prevents hypoglycaemia and increasesfructose 2,6-biphosphate in TNF (-/-) mice administeredendotoxin, MIF is argued to act independently of TNF[42]. The relative importance and interaction of HIF-1 and

MIF in this context have not yet been examined, but theseactivities of TNF and MIF nevertheless stress that anunderstanding of malarial disease needs a broader visionthan simple vascular occlusion.

Metabolic acidosis

Metabolic acidosis is not a disease, but a symptom of aserious underlying process. As recently summarized [274],metabolic acidosis is defined as a decrease in blood pHsecondary to a decrease in the bicarbonate concentration.

The decrease in bicarbonate concentration may be sec-ondary to an excess of acids that will consume bicarbo-nates, reflecting the open character of this buffering

system, or to a loss of bicarbonates through the digestiveor the renal route. These authors [274] note that impor-tance of the metabolic acidosis may be appreciated bymeasuring the change in bicarbonate concentration fromthe normal, or the base excess, which gives a better assess-ment of the acid load because it takes into account thebuffering of the non-bicarbonate systems. Despite theimpression that much malaria literature gives, metabolicacidosis is not unique to this disease, being seen in viral,rickettsial and bacterial infections [240] as well as acutegastroenteritis, where its prevalence is higher than inmalaria [275].

Lactate: a cause, marker, or neither in the acidosis of malaria orsepsis?

The notion persists in some malaria circles that excess lac-tate accumulation causes the metabolic acidosis that cor-relates with a poor clinical outcome, and therefore

warrants therapeutic reduction [4,276]. Accordingly,these authors argue that lowering lactate levels withsodium dichloroacetate (DCA), an inhibitor of pyruvatedehydrogenate kinase, would be sufficient to amelioratethe metabolic acidosis of falciparum malaria [4,276,277].In our view the current literature on systemic inflamma-

Table 2:

Influenzaencephalopathy

Cerebralmalaria

seizures/coma after high grade

fever

+ +

metabolic acidosis + +

hyperlactataemia + +

serum TNF, IL-6, sTNFRI up + +

serum nitrite/nitrate up + +

CSF TNF, IL-6, sTNFRI up + +

multiple organ failure, sequelae + +

thrombocytopaenia + +

damage to vascular endothelialcells

+ +

brain oedema/damage to BBB + +

apoptosis in neurons/glial cells + +

evidence of active caspase-3(brain)

+ +

caspase-cleaved PARP (brain) + +
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


16/32



tory diseases does not support this. To successfully treatmetabolic acidosis DCA would need to increase oxygenconsumption so that oxidative phosphorylation takesover from anaerobic glycolysis as the dominant ATP pro-ducer, and the protons generated when ATP is resynthe-

sized from ADP and inorganic phosphate are re-consumed, increasing pH to normal levels [278].

A recent editorial in Critical Care Medicine [279] haslucidly summarized the key points of the mechanism ofmetabolic acidosis in sepsis, a condition that shares sys-temic inflammation and a range of its consequences withsevere malaria (Table 2). These authors do not accept thatlactate is the cause of the acidosis associated with hypoxia.Instead, they note the evidence that cellular acidosis dur-ing hypoxia, be it from limited oxygen supply or utilisa-tion, arises from the hydrolysis of non-mitochondrial

ATP. Every time a molecule of ATP undergoes hydrolysis,

a proton is released. When oxygen is readily available, theproducts of this reaction, including protons, are recycledby mitochondria, and pH does not change. Duringhypoxia, however, the mitochondrial turnover rate dropsbelow the rate of ATP hydrolysis, so protons are being pro-duced faster than they can be recycled, and intracellularpH falls once buffering capacity is exceeded. Since thesereactions are independent of lactate levels, merely reduc-ing the level of this anion can therefore no more beexpected to increase survival rate in falciparum malariathan it did in sepsis [280]. Indeed, it could in theory harmcomatose patients, since there is evidence that lactatehelps brain tissue survive hypoxic and hypoglycaemic epi-

sodes [281-283], and the lactate shuttle is proving to behow astrocytes protect neurons from metabolic stress[284]. Moreover, infusing enough lactate into patients

with severe sepsis to cause hyperlactataemia did not causeacidosis, but an alkalosis [285].

Even when considerable lactate is generated

Documents

A Consequence of Inflammatory Cytokine