[page 8↓]

1.  Introduction


The family of mycobacteria comprises pathogens and apathogenic environmental bacteria [1,2]. Mycobacteria are unusual among bacteria since they have an enormously thick, hydrophobic cell wall which e.g. prevents desiccation. Numerous mycobacteria are harmless and useful because they degrade organic matter in soil. Better known are, however, the few human pathogenic mycobacteria which cause tuberculosis (Mycobacterium tuberculosis, Mycobacterium africanum, Mycobacterium bovis) and leprosy (Mycobacterium leprae). Discovery of M. tuberculosis as an etiologic agent was reported to the “Physiologische Gesellschaft” in Berlin on March 24th, 1882 by Robert Koch. During his lecture on the “Ätiologie der Tuberkulose” Koch pointed out that it was a specific staining procedure that visualized characteristic, so far unknown bacteria in tuberculous affected organs [3]. The staining was performed with alkaline methylene blue and bismarck brown. Hence, M. tuberculosis as causal agent of tuberculosis was identified. At about the same time of Robert Koch’s discovery, Paul Clemens von Baumgarten brightened bacilli with caustic soda in caesous tissue and saw the bacteria under the microscope. Baumgarten’s publication came out a few weeks after Robert Koch’s lecture in the Charité in Berlin[4]. The staining procedures were modified first, by Paul Ehrlich, and subsequently by Franz Ziehl and Adolf Neelsen, to the use of a mixture of acids and aniline dye, today known as “Ziehl-Neelsen-staining” [5,6,7]. Therefore, the mycobacterial cell wall and its staining attributes were of notable importance right from the beginning.

From a public health standpoint, M. tuberculosis is of utmost importance because it kills more human beings than any other bacterial species [page 9↓](www.who.int/mediacentre/factsheets/who104/en/print.html). The World Health Organization (WHO) declares that approximately 2 million people die of tuberculosis each year. It is estimated that between 2002 and 2020, approximately 1000 million people will be newly infected, over 150 million people will develop disease, and 36 million will die of tuberculosis, if control is not improved. HIV and tuberculosis form a lethal combination, each speeding up the other's progression. Because HIV weakens the immune system, an HIV-positive individual who is infected with M. tuberculosis is many times more likely to develop disease than someone infected with M. tuberculosis who is HIV-negative. Tuberculosis is the leading cause of death among people who are HIV-positive with about 11% of AIDS deaths worldwide. In Africa, HIV is the single most important factor determining the increased incidence of tuberculosis in the past 10 years. By inconsistent or partial treatment, drug resistant bacilli have emerged. A particularly dangerous form of drug-resistant tuberculosis are the multidrug-resistant (MDR) strains causing MDR-tuberculosis, which is defined as the disease caused by bacilli resistant to at least Isoniazid and rifampicin, the two most powerful anti-tuberculosis drugs.

1.1. Koch’s postulate and Koch’s molecular postulate


Robert Koch’s lecture on the “ Ätiologie der Tuberkulose ” provided an experimental methodology and an intellectual basis for acquiring knowledge about the cause of tuberculosis. Koch’s postulate was of great importance:

To prove that tuberculosis was caused by the invasion, growth and multiplication of the bacilli, it was necessary to:

  1. solate the bacilli from the body,
  2. grow them in pure culture, and
  3. by administering the isolated bacilli to animals, reproduce similar moribund conditions.


Similarly, the knowledge of the molecular bases (the genotypes) for the phenotypes which allow M. tuberculosis to cause disease and escape therapies is essential in developing new effective treatment strategies. The conceptual methodology by which we acquire knowledge about what genotype is responsible for a particular phenotype is called Koch’s molecular postulate [8]:

To prove that a phenotype in a mutant bacterium, such as drug resistance or virulence, is caused by a specific genotype, it is necessary to:

  1. Isolate a mutant with a defined altered phenotype,
  2. clone the genotype from the mutant, and
  3. by introducing the cloned genotype into a wildtype bacterium, reproduce the same phenotype of the mutant bacterium.


Based on this principle, a large number of genetic functional analyses regarding a variety of virulence and other factors in different bacteria have been achieved. Technological progress paved the way for the generation of mutant libraries and targeted gene deletions, as well as overproduction of selected genes. M. tuberculosis has been one of the most challenging and refractory organisms in this process. Nevertheless, there exists a set of tools to manipulate these bacilli [9], e.g. the[page 11↓] sophisticated TM4-phage based knockout technology [9].

1.2.  Mycobacterium bovis BCG


Soon after Robert Koch’s discovery of M. tuberculosis, innumerable attempts toward vaccine development began. On December 28th, 1908, the French bacteriologists Albert C. Calmette and Guérin notified a loss of virulence of M. bovis when cultured in bile containing media. These scientists passaged M. bovis over a period of 13 years in a bile-glycerin-medium thereby developing an attenuated strain that, when used as a vaccine, provided protection to high risk groups, especially newborns of tuberculous mothers. This vaccine strain was called Mycobacterium bovis BCG. It was introduced into the clinic in 1921 and is one of the oldest applications of vaccination. Nevertheless, there were severe complications [10] and later it was proven that M. bovis BCG fails to provide satisfying protection in adults against reactivated pulmonary tuberculosis [11,12]. In the Federal Republic of Germany the routine vaccination was discontinued in 1985 and has been offered since 1987 only upon request. Since there is no better anti-tuberculous vaccine than M. bovis BCG available so far, it still is the golden standard to which new candidates have to be compared in animal models.

1.3. Immune response to M. tuberculosis


The immune system is divided into two categories: innate and adaptive. Innate immunity refers to nonspecific defense mechanisms that come into play immediately or within hours of an antigen’s appearance in the body. These mechanisms include physical barriers such as skin, effector molecules in body fluids such as mucosal [page 12↓]secretions and blood, and immune cells of the myeloid lineage. The innate immune response is activated by chemical properties of the antigen. Adaptive immunity, on the other hand, refers to antigen-specific immune responses. The adaptive immune response is more complex than the innate. The antigen first must be processed and presented. Once an antigen has been recognized, the adaptive immune system creates a set of immune cells specifically designed to attack that antigen or the antigen bearing pathogen within days or weeks. The adaptive immunity includes a “memory” that renders future responses against a specific antigen more efficient. Innate and adaptive immunity are closely inter-connected. Macrophages and dendritic cells, the primary cell types involved in the innate immune response to mycobacteria, play a crucial role in the initiation of adaptive immunity. For example, infected macrophages are recognized by T lymphocytes through their major histocompatibility complex (MHC) presentation of mycobacterial antigens.
Tuberculosis is caused by airborne infection. Inhaled droplets containing low numbers of bacteria are taken up by alveolar macrophages. Spontaneous healing cannot be measured nor excluded but is unlikely to occur to a significant degree. The vast majority, about 90 % of infected individuals, does not develop acute disease and stays latently infected. Patients with a compromised immune system, for example in the case of HIV infection, develop acute disease directly after primary infection [13]. Within 1 year more than 10 % of infected individuals develop disease. M. tuberculosis resides in early phagosomes and blocks phagosome maturation including phagolysosome formation [14,15,16,17]. However, the maturation arrest is incomplete and some bacteria are killed or at least impaired in replication through antibacterial effectors including reactive oxygen and nitrogen intermediates [18]. In addition, iron restriction is an important mechanism of innate immunity to control the infection [19].
[page 13↓]Bacterial containment is focused on the granulomatous lesion, where different T cell populations participate in the protective immune response. These include i) CD4+ T cells recognizing antigenic peptides in the context of gene products encoded by MHC class II, ii) CD8 T cells recognizing antigenic peptides in the context of MHC class I, iii) γδ T cells recognizing unusual antigenic ligands independent of specialized presentation molecules – notably phospholigands, and iv) CD1 restricted T cells recognizing glycolipids abundant in the mycobacterial cell walls presented by the CD1 molecules [20,21]. Upon infection/reactivation antigen specific T cells produce interferon γ (IFNγ) which synergizes with tumor necrosis factor α (TNFα) in activating macrophages. At least some of the CD8 T cells, γδ T cells, and CD1 restricted T cells secrete perforin and granulysin, thereby directly killing mycobacteria within macrophages (Fig. 1) [22]. Most of this knowledge was obtained from experiments with mice. The immune response of the mouse is well understood, and a large variety of mouse mutants with defined immunodeficiencies are available. Furthermore, the function of IFNγ, IL12, TNFα, or CD4 T cells is similar in mouse and human [23]. Although, there are significant differences between the human and murine immune system, experimental animals are critical to gain insight into general mechanisms underlying natural resistance, and acquisition of a protective immune response.


[page 14↓]

Fig. 1 Scheme of the course of events following contagion with M. tuberculosis. Acute disease does only develop for a relatively small subset of immunocompromised individuals. Depicted are the major effector mechanisms of macrophages and the most important T cell populations. (Reprinted from Nature Reviews Immunology [20]).

1.4. The mycobacterial cell wall


The mycobacterial cell wall is impressively thick and unique in its complex composition [24]. Among the strategies by which M. tuberculosis has adapted to the environmental conditions in macrophages, the resistance imparted by its cell wall is one of the most striking. The cell envelope is extremely hydrophobic and forms an exceptionally strong permeability barrier, rendering mycobacteria naturally resistant to a wide variety of antimicrobial agents. This is due to the unique structure of the mycobacterial cell wall and the presence of long fatty acids, the mycolic acids [24]. Channel forming proteins which are functionally similar to the well known porins of gram negative bacteria have been demonstrated in Mycobacterium chelonae [25] [page 15↓]and M. smegmatis [26,27,28], revealing how hydrophilic molecules can pass through the hydrophobic cell wall. The core unit of the envelope consists of peptidoglycan connected to arabinogalactan (AG) which is covalently linked to mycolic acids, thus forming the mycolyl-AG-peptidoglycan complex (MAPc) (Fig. 2). Cell wall synthesis can be divided into 3 separate stages which occur in distinct subcellular compartments:

Fig. 2 Scheme of the mycobacterial cell wall structure. The mycolyl-arabinogalactan-peptidoglycan complex forms the core of the robust bacterial envelope.


cytoplasm, membrane, and the cell wall itself (Fig. 3). Peptidoglycan synthesis is initiated with Uridine 5’-Diphospho-N-Acetylmuramic Acid (UDP-MurNAc) derived from Uridine 5’-Diphospho-DN-AcetylGlucosamine (UDP-GlcNAc) and phosphoenylpyruvate. Five amino acids are linked to UDP-MurNAc resulting in synthesis of UDP-MurNAc-pentapeptide, also known as Park’s nucleotide [29]. UDP[page 16↓]-MurNAc-pentapeptide is linked via a phosphodiester to an undecaprenyl pyrophosphate carrier molecule (C55-PP) constituting C55-PP-MurNAc-pentapeptide, or lipid I. A GlcNAc residue is subsequently added to form lipid II, which is thought to be translocated across the cytoplasmic membrane and serves as substrate for the assembly of peptidoglycan [29]. Generation of undecaprenyl monophosphate (C55-P) from undecaprenyl (C55) by an

Fig. 3 Model of peptidoglycan synthesis. Precursors are produced in the cytoplasm subsequently coupled to a undecaprenyl monophosphate, the carrier lipid, and modified. After translocation to the outer face, peptidoglycan is assembled and incorporated into the cell wall.


undecaprenyl-phosphokinase (Upk) probably represents a critical step in providing mycobacteria with a sufficient amount of peptidoglycan-precursor carrier molecules. Bacitracin is an antibiotic which binds tightly to C55-PP and blocks C55-P-recycling, in this way also reducing the amount of carrier molecules.


[page 17↓]

1.5.  Mycobacterium smegmatis


To study mycobacterial genetics of tuberculosis, the fast growing, environmental Mycobacterium smegmatis is widely used as a model organism. E.g., it was used to understand the effect of single M. tuberculosis genes on survival inside human macrophages [30]. Another potential use is to serve as surrogate for M. tuberculosis in disinfectant tests. Discovery of M. smegmatis was made 2 years after Robert Koch’s lecture on the “aetiology of tuberculosis”. In November 1884, Lustgarten was first to describe a bacillus with staining appearance of tubercle bacilli in syphilitic chancres and gummae [31]. Soon thereafter Alvarez and Tavel identified microorganisms similar to those in normal genital secretions (smegma) [32]. This kind of growth, as bacterial communities on a surface, is defined as a biofilm. Various gram-negative and gram-positive bacteria as well as fungi show two forms of growth: planktonic and, as a step of microbial development, in biofilms [33]. M. smegmatis and other non-tuberculous mycobacteria such as Mycobacterium fortuitum and Mycobacterium marinum live and grow planktonic or as biofilms [34].Biofilms support resistance to antimicrobial chemotherapy [35] and play a role in contamination of clinical and industrial settings [36]. M. smegmatis has also been described in wound infections, endocarditis, and osteomyelitis [37]. Nevertheless, it is considered as an apathogenic fast growing mycobacterium. Therefore, M. smegmatis remains an attractive model organism to study mycobacterial metabolism or, specifically, cell wall synthesis as investigated in this study.

1.6. Aim of the study


As mentioned, the cell envelope is a key to understanding mycobacteria and especially their resistance. Understanding initial processing, transport and assembly of peptidoglycan [page 18↓]precursors, therefore is key to understanding the mycobacterial cell wall. In a first approach, the hypothesis, that gene products which are relevant in this context in other bacterial species, have homologues in mycobacteria with comparable function, was followed. Koch’s molecular postulates were used to test this hypothesis as follows:

1.) Isolate a mutant with a defined altered phenotype.

Attention was drawn to undecaprenyl phosphokinase, which is involved in peptidoglycan synthesis, and mediates resistance to the antibiotic bacitracin in the case of Escherichia coli, Staphylococcus aureus, and Streptococcus pneumoniae. Furthermore, the altered phenotype is characterized by decreased virulence for Staphylococcus aureus and Streptococcus pneumoniae [38,39].

2.) Clone the genotype from the mutant

The sequences of E. coli, S. aureus, and S. pneumoniae undecaprenyl phosphokinases are known. The homologous genes of M. tuberculosis, M. bovis BCG and M. smegmatis had to be identified, and appropriate knockout constructs had to be cloned.

3.) Introduce the cloned genotype into a wildtype bacterium, and reproduce the same phenotype of the mutant bacterium .

Subsequently, the knockout constructs had to be used to introduce the deletions into the selected wildtype mycobacteria, followed by verification that a phenotype comparable to that [page 19↓] [page 20↓]of mutants in other species can be reproduced.
An additional aim was the more detailed characterization of Δupk mutants in different mycobacteria. This included determination of influence on peptidoglycan synthesis, bacitracin resistance, biofilm formation in the case of M. smegmatis, and virulence and persistence in the case of M. tuberculosis and M. bovis BCG. This was done in order to:

determine steps in cell wall synthesis which represent potential targets for drug development, and

generate an attenuated new vaccine candidate.


© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 3.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML generated:
15.07.2004