151

The Genetics of Host Resistance and Susceptibility to Tuberculosis

PETER DAVIESa AND JOHN GRANGEb

aConsultant Respiratory Physician, Cardiothoracic Centre, Liverpool L14 3PE, U.K.bDepartment of Microbiology, University College, London, UK

ABSTRACT: The study of human genomics has the potential to aid our under-standing of the interindividual and interpopulation differences in susceptibilityto tuberculosis. Resistance to infection is affected by the ability of macrophagesto phagocytose and destroy the bacilli. Several genes are involved in this pro-cess, and two have been the focus of recent interest: the natural resistance-associated protein (NRAMP1) gene and the genes coding for the vitamin Dreceptor. Susceptibility genes have also been discovered—for example, one onthe X chromosome that may explain the increased susceptibility of males to tu-berculosis. Studies have also focused on the variations in virulence of the bacil-lus in both its drug-susceptible and drug-resistant forms. These mechanismsmust be understood in order to prevent, or combat, the emergence of a viru-lent, multidrug-resistant form of the bacillus that would be uncontrollable bymeans of today’s treatment strategies.

KEYWORDS: tuberculosis; human genomics; vitamin D; X chromosome; multi-drug-resistant tuberculosis; immune function

INTRODUCTION

One of the great mysteries of tuberculosis is that 90% of people infected with theorganism never develop the disease. If even part of the reason for this is attributableto the genetic characteristics of these “resistant” individuals, then a careful study ofthe human genome should provide important information to help control and eveneliminate this disease in the individual human host and the human population.

Within the group of people susceptible to disease, those least able to controlgrowth of bacteria within their tissues run the highest risk of developing acquireddrug-resistant disease, because the numbers of bacteria undergoing natural mutationto drug resistance will potentially be greater. The study of the genetic control of tu-berculosis, to understand how we resist tuberculosis, is therefore relevant to the un-derstanding of why drug resistance may arise in a human population.

Address for correspondence: Dr. Peter Davies, Cardiothoracic Centre, Thomas Drive, Liver-pool L14 3PE, U.K. Voice: 44 151 293 2392; fax: 44 151 228 7688.

P.D.O.Davies@liverpool.ac.uk

152 ANNALS NEW YORK ACADEMY OF SCIENCES

GENETIC BASIS OF VARIATIONS IN RESISTANCE

Many studies, notably those on monozygotic and dizygotic twins, indicate thatinherited genetic factors play a key role in determining susceptibility and resistanceto overt tuberculosis following infection.1 The immune interactions between the hostand the highly complex molecular structure of the tubercle bacillus is multifactorial.From the standpoint of the host, resistance to infection is affected by the ability ofmacrophages to phagocytose and destroy the bacilli; and this, in turn, is affected byclass II HLA-determined antigen presentation, the activation of macrophages by T-cell-derived cytokines and vitamin D,2,3 and granuloma formation. Several of thegenes involved in these processes have been identified, notably the HLA-DR2 andHLA-DQB1 loci, that determine which mycobacterial antigens are presented tohelper T cells.4,5 Non-HLA genes are also involved, and two in particular have beenthe focus of recent interest. One of these is the so-called natural resistance-associat-ed macrophage protein (NRAMP1) gene.6 This is the human equivalent of a well-described gene in the mouse responsible for resistance to intracellular parasitesincluding Leishmania, Salmonella, and bacille Calmette-Guérin (BCG). Its functionin determining human resistance to tuberculosis is unknown, although there is someevidence that it regulates the concentration of iron and other cations in the phago-somes of the macrophage.

The other non-HLA gene that appears to be involved in resistance to tuberculosisare those coding for the vitamin D receptor.2 The importance of vitamin D in resis-tance to tuberculosis has been known since the 18th century, and cod liver oil andexposure to sunlight were once widely used treatments for this disease, with appar-ent benefit.7 Interest reemerged during investigations on the cause, previously inad-equately explained, of the relatively high incidence of nonrespiratory manifestationsof tuberculosis in patients of Indian subcontinent ethnic origin living in the UnitedKingdom. The fact that a relatively high incidence of nonrespiratory tuberculosis oc-curs in AIDS patients led to the hypothesis that immigration might be associatedwith some form of acquired immunodeficiency.8 A likely cause of this is a lowering,up to a 10-fold lowering in some cases, of serum vitamin D (25 hydroxycholecalcif-erol) levels in immigrants from tropical countries after arrival in the cold and cloudyUnited Kingdom.9 This vitamin is essential for the full expression of macrophage ac-tivation, a key component of protective immunity in tuberculosis.10 It is thereforepostulated that people may be infected and develop latent tuberculosis in their coun-tries of origin and proceed to active disease after their vitamin D levels fall in theirnew home country.

The ability of vitamin D to activate macrophages, and thereby to affect the out-come of infection by the tubercle bacillus, is genetically determined. Studies on apredominantly vegetarian population of Gujarati Asians living in west London sug-gested that the expression of genetically determined differences in vitamin D func-tion are accentuated under conditions causing a deficiency of this vitamin. Thus thecombination of the TT/Tt VDR genotype and vitamin D deficiency, and of the ff gen-otype and undetectable serum vitamin D levels, strongly predisposed these personsto active tuberculosis.3

A number of other genes affecting immune function have been associated withresistance to tuberculosis,11 including genes determining the allotypes of the Km1light chain immunoglobulin,12 and of haptoglobin.13,14 Whether these genes are of

153DAVIES & GRANGE: GENETICS OF REISTANCE AND SUSCEPTIBILITY

direct functional importance or are merely linked to resistance-determining genes re-mains to be determined. Other genes of putative relevance are those determining thefunction of various cytokines involved in protective immunity.15 Such specific ge-netic defects are rare; and, in general, susceptibility and resistance to tuberculosis isthe result of a large number of genes inherited in a complex way, as clearly demon-strated in family-based linkage studies and population-based case-control studies.11

The principal determinants of susceptibility and resistance to tuberculosis that havebeen shown to be affected by genotypic variation are listed in TABLE 1.

SUSCEPTIBILITY GENES

Resistance to tuberculosis is, at least in part, determined by genetic factors: cer-tain genes are responsible for susceptibility to the disease. A major susceptibility al-lele may arbitrarily be defined as one with prevalent susceptibility variantsdetermining a relative risk of disease of greater than threefold. A high prevalence ofthis allele within the disease population, usually of greater than 10%, will also berequired to show susceptibility. In general, a genetic effect of a particular size mayresult from a large number of minor susceptibility genes (polygenes) or from a fewmajor genes with polygenes. A large number of candidate genes, for example thoseaffecting the synthesis of the cytokines tumor necrosis factor (TNFα) and INF-γ, thatmight have been expected to show association with tuberculosis have not done so.

TECHNIQUES OF GENE DETECTION

There are two methods for identifying susceptibility genes, and each has its mer-its and weaknesses. Linkage studies are family based, and data collection may provedifficult. They also tend to have a low power, so that moderate effects can be missed.On the other hand, as they are systematic and comprehensive, they are able to detect

TABLE 1. Factors involved in susceptibility or resistance to tuberculosis and subjectto genetically determined variation

Factor or gene Function or mode of action

Natural resistance-associated macrophage protein (NRAMP 1)

? Regulation of phagosome cation levels

HLA Class II Antigen presentation

Vitamin D receptor Macrophage activation

Locus on chromosome 15 Unknown

Locus on X chromosome Unknown: may account for higher rates of tuberculosis in males

Haptoglobin ? Regulator of lymphocyte function

Km1 immunoglobulin allotypes ? Related to autoimmune tissue damage

Various cytokines T-cell maturation and patterns of immune reactivity

154 ANNALS NEW YORK ACADEMY OF SCIENCES

all major genes. Data is easier to collect in case-control association studies; and, be-cause of their higher power, they can detect moderate effects. As they are based onthe candidate gene approach, however, they can miss major genes. Some 300 mark-ers are required to screen the entire human genome by linkage analysis compared toapproximately 100,000 markers required to screen the human genome by associa-tion. There are about 200 genes in the region of a linkage.

An extensive genome-wide study in the Gambia and South Africa, based on fam-ilies with two or more siblings developing tuberculosis, revealed two chromosomalregions that linked to susceptibility to clinical tuberculosis.11 An independent asso-ciation-based analysis using microsatellite markers supported the presence of sus-ceptibility genes on chromosome 15 and the X chromosome. The finding of asusceptibility locus on the X chromosome may explain the widely documented in-creased susceptibility of males to tuberculosis. In the Gambian study, 70% of the tu-berculosis patients were male.

THE MULTIFACTORIAL NATURE OF SUSCEPTIBILITY TOTUBERCULOSIS

It may therefore be concluded that the genetically determined components of theimmune responses to tuberculosis and leprosy16 are principally the results of the cu-mulative impact of a very large number of minor genes, but some major loci existand may be mapped by linkage analysis of affected sibling pairs. From this it maybe concluded that the outcome of infection depends on a highly complex interactionbetween the host immune system and the numerous molecules comprising the tuber-cle bacillus. Just as the innate and acquired host determinants of resistance and sus-ceptibility are complex and multifactorial, so are the determinants of virulence of thetubercle bacillus. Many putative determinants of mycobacterial virulence have beendescribed, and these interact with various components of the immune system. Theseinclude factors facilitating uptake of bacilli by macrophages and other cells and sur-vival within macrophages and more recently discovered factors affecting bacterialdormancy and the phenomenon of latency.

Mycobacterial virulence factors determine the chance that an infection will pro-ceed to a recognizable change in immune function, an immunologically effectivecontact, usually detected as the development of tuberculin positivity, and the chanceof infection progressing to clinically overt disease—the disease ratio. Determinationof both in humans is notoriously difficult, because there is no way of detecting in-fection that does not lead to immunological changes, and the nature of the infectingbacillus leading to the development of tuberculin positivity but not overt disease can-not be determined. In this context, it has been suggested that the serial mutationscausing a drug-susceptible tubercle bacillus to become multidrug resistant would af-fect, possibly reduce, its “fitness” or virulence.17 If this is indeed the case, a low dis-ease ratio would lead to the eventual decline in prevalence of initial cases ofmultidrug-resistant tuberculosis, provided that the emergence of new cases of ac-quired resistance were prevented by strict application of the World Health Organiza-tion DOTS (directly observed treastment, short term) strategy. If, on the other hand,multidrug-resistant bacilli retain full virulence, then specific, albeit very costly, con-trol measures such as the “DOTS-plus” strategy would be essential. One recent study

155DAVIES & GRANGE: GENETICS OF REISTANCE AND SUSCEPTIBILITY

suggests that the transmissibility and disease ratio of drug-susceptible andmultidrug-resistant tuberculosis is of a similar order.18 (In this respect, early find-ings that isoniazid-resistant strains were of low virulence for the guinea pig led tothe dangerously false assumption that virulence for humans was likewise reduced.The ensuing use of isoniazid monotherapy, notably in India, for cases of overt tuber-culosis generated a problem of isoniazid resistance that is with us today.) Clearly,further epidemiological and molecular microbiological work on this important issueis required in order that optimum disease control strategies may be planned.

CONCLUSIONS

The study of human genomics has a great potential to aid our understanding ofthe interindividual and interpopulation differences in susceptibility to tuberculosis.It can also be used to identify the effects of millennia of contact with M. tuberculosison the evolution of our genome. The genomes of both Homo sapiens and M. tuber-culosis have been mapped, and both remarkable achievements should help us toidentify all the immunological, molecular, and biochemical pathways involved insusceptibility and resistance to tuberculosis and thereby aid the development of nov-el vaccines, diagnostic tests and therapies—both drugs and immunotherapeuticagents.

Over millennia, the human genome has evolved to afford variable though general-ly good protection against a fairly homogeneous population of tubercle bacilli. Onlyfor the last few decades have genetically altered multidrug-resistant bacilli been se-lected for, producing unpredictable interactions with our immune systems. Animaland in vitro cell systems, despite inherent weaknesses and dubious relevance to hu-man disease, suggest that tubercle bacilli are divisible into “wimpy” and “burly” vari-ants.19 Are those that have mutated to multidrug resistance, of genetic necessity,always among the “wimps,” or could “burly” resistant variants arise and cause seri-ous, spreading epidemics uncontrollable by the strategies in place today? Answers tothese questions are urgently required, before a potential human health disaster strikes.

ACKNOWLEDGMENT

We would like to thank Professor Adrian Hill, who gave the original talk at themeeting, for his helpful suggestions in compiling this paper.

REFERENCES

1. COMSTOCK, G.W. 1978. Tuberculosis in twins: a re-analysis of the Prophit survey. Am.Rev. Respir. Dis. 117: 621–624.

2. BELLAMY, R., C. RUWENDE, T. CORRAH, et al. 1999. Tuberculosis and chronic hepatitisB virus infection in Africans and variation in the vitamin D receptor gene. J. Infect.Dis. 179: 721–724.

3. WILKINSON, R.J., M. LLEWELYM, Z. TOOSSI, et al. 2000. Influence of vitamin D defi-ciency and vitamin D receptor polymorphisms on tuberculosis among GujaratiAsians in west London: a case-control study. Lancet 355: 618–521.

156 ANNALS NEW YORK ACADEMY OF SCIENCES

4. BOTHAMLEY, G.H., J.S. BECK, G.M. SCHREUDER, et al. 1989. Association of tuberculo-sis and M. tuberculosis-specific antibody levels with HLA. J. Infect. Dis. 159: 549–55.

5. GOLDFELD, A.E., J.C. DELGADO, S. THIM, et al. 1998. Association of an HLA-DQ allelewith clinical tuberculosis. J. Am. Med. Assoc. 279: 226–228.

6. BELLAMY, R., C.R. RUWENDE, T. CORRAH, et al. 1998. Variations in the NRAMP-1gene and susceptibility to tuberculosis in West Africans. N. Engl. J. Med. 338: 640–444.

7. 1849. Report: Cod-liver oil treatment of tuberculosis. Brompton Hospital Records1849: 38.

8. DAVIES, P.D.O. 1985. A possible link between vitamin D deficiency and impaired hostdefence to Mycobacterium tuberculosis. Tubercle 66: 301–306.

9. DAVIES, P.D.O. 1989. Vitamin D and tuberculosis. Am. Rev. Respir. Dis. 139: 1571.10. ROOK, G.A.W. 1988. The role of vitamin D in tuberculosis. Am. Rev. Respir. Dis.

138: 768–770.11. BELLAMY, R., N. BEYERS, K.P. MCADAM, et al. 2000. Genetic susceptibility to tuber-

culosis in Africans: a genome-wide scan. Proc. Natl. Acad. Sci. USA 97: 8005–8009.

12. GIBSON, J.A., J.M. GRANGE, J.S. BECK & T. KARDJITO. 1987. An association betweenKm1 immunoglobulin allotype and pulmonary tuberculosis in Indonesia. Exp.Clin. Immunogenet. 4: 129–135.

13. SHENDEROVA, R.J. & N.A. OSINA. 1991. Syvorotochenye sistemy krovi u bol’nykhtuberkulezom [The blood serum systems in patients with tuberculosis]. Probl.Tuberk. 3: 54–56. (In Russian with English summary)

14. GRANGE, J.M., T. KARDJITO, J.S. BECK, et al. 1985. Haptoglobin: an immunoregula-tory role in tuberculosis? Tubercle 66: 41–47.

15. BELLAMY, R., C. RUWENDE, T. CORRAH, et al. 1998. Assessment of the interleukin 1gene cluster and other candidate gene polymorphisms in host susceptibility totuberculosis. Tuberc. Lung Dis. 79: 83–89.

16. SIDDIQUI, M.R., S. MELSNER, K. TOSH, et al. 2001. A major susceptibility locus forleprosy in India maps to chromosome 10p13. (letter) Nature Genet. 27: 439–441.

17. REMUS, N., J. REICHENBACH, C. PICARD, et al. 2001. Impaired interferon gamma-mediated immunity and susceptibility to mycobacterial infection in childhood.Paediatr. Res. 50: 8–13.

18. DYE, C. & B.G. WILLIAMS. 2000. Criteria for the control of drug-resistant tuberculo-sis. Proc. Natl. Acad. Sci. USA 97: 8180–8185.

19. SMALL, P.M. 1999. Tuberculosis in the 21st century: DOTS and SPOTS. Int. J.Tuberc. Lung Dis. 3: 949–955.