- Email: [email protected]
Change in colony morphology influences the virulence as well as the biochemical properties of theMycobacterium aviumcomplex
Microbial Pathogenesis 1998; 25: 203–214 Article No. mi980227
MICROBIAL PATHOGENESIS
Change in colony morphology influences the virulence as well as the biochemical properties of the Mycobacterium avium complex Rita G. Kansal, Ricardo Gomez-Flores & Reeta T. Mehta∗ Department of Bioimmunotherapy, The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030 U.S.A. (Received October 7, 1997; accepted in revised form August 11, 1998)
Factors that influence colony morphology are of crucial importance for drug development as well as for understanding the virulence of Mycobacterium avium complex (MAC) strains. The MAC 101 strain used in the present study grows as smooth transparent (SmT) colonies that tend to become opaque and pigmented when incubated for long periods of time. However, when MAC was passaged in animals, two types of colonies were recovered. The new rough transparent (RgT) colony morphology appeared more flat and transparent, having a central spot, irregular edges at times, and a dry, granular appearance like that of the rough mutants. In animal studies, the RgT bacilli multiplied at a much faster rate than that of the SmT bacilli, causing 60–80% mortality compared with the 10% mortality observed in mice infected with SmT. In vitro studies indicated that the SmT MAC did not grow and multiply as well in resident peritoneal macrophages as the RgT MAC did. The two morphotypes did not differ in their growth rates in vitro but the RgT MAC failed to reduce dimethylthiazol-diphenyltetrazolium bromide (MTT), alamar blue and neutral red, suggesting that there might be significant changes in the cell wall or elsewhere causing changes in cellular permeability. These two morphotypes could serve as models for studying the biochemical markers or the identification of factors responsible for the virulence of the MAC. 1998 Academic Press
Key words: Mycobacterium avium complex, colony morphology, virulence, growth rate, biochemical properties, MTT reduction.
Introduction ∗ Author for correspondence at: Department of Bioimmunotherapy, Box 60, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030, U.S.A. 0882–4010/98/100203+12 $30.00/0
The Mycobacterium avium complex (MAC) belongs to a group of bacteria known as atypical mycobacteria. Infection caused by the MAC is one of the most common opportunistic infections 1998 Academic Press
204
afflicting patients with advanced acquired immunodeficiency syndrome (AIDS) [1–4]. There is compelling evidence that this infection in humans contributes to a decrease in both quality of life and survival [5, 6]; its effective treatment does prolong survival as well as quality of life [7]. The pathogenesis of the MAC is not well understood. However, the colony morphology of MAC, distinguished into three groups (SmT, SmOp and RgT) has been associated with virulence in chickens and mice [8–11]. Although factors that influence colony morphology are of crucial importance, existing studies do not provide conclusive information; one of the important factors contributing to this is the high rate of interconversions of these colonies [12]. It has been reported that the virulence of the infecting strain of MAC 101 increases markedly following passage in animals [13], but no comparison with colony morphology is available. The MAC 101 strain used in the present study grows as smooth transparent (SmT) colonies that tend to become opaque and pigmented after prolonged incubation. However, when passaged in animals, a new colony morphotype emerged that appeared flatter with a central spot, irregular edges at times and a dry granular appearance like that of the rough transparent (RgT) mutants. The bacilli recovered from infected organs formed RgT type colonies that increased in number as the time of infection increased. This morphotype could be maintained in macrophages without passage in the animals. In an attempt to study the underlying mechanisms that control colony morphology, the two morphotypes were isolated and studied with respect to their growth, metabolic status, and virulence.
Results Morphological appearance MAC 101 grows as SmT colonies on 7H11 agar that tend to become opaque and pigmented after prolonged incubation. However, when passaged in animals, two colony types could be recovered. The new colony morphotype (RgT) appeared flatter with irregular edges at times and was dry in appearance like the rough mutants. These colonies look similar and are difficult to distinguish until 10 days after agar plating of the culture; as they grow older, the SmT colonies
R. G. Kansal et al.
Figure 1. Typical colony morphology of 21-day-old SmT and RgT colonies under simple microscopy (×40).
become more opaque and pigmented than the RgT colonies which remain transparent and white. Fig. 1 shows the typical colony morphology of 21 day old SmT and RgT colonies under light microscopy (×100). These two morphotypes were then isolated and grown separately in Middlebrook 7H9 broth (MBB) for further studies. Broth cultures of the SmT and RgT MAC showed visible differences in turbidity (the RgT MAC had much less turbidity than the SmT MAC) with no differences in the number of colony forming units (cfu) per ml [Fig. 2(a)].
Mycobacterial growth determination The growth of the SmT and RgT morphotypes in MBB from day 1 to day 8 yielded interesting results. Whereas growth in terms of viable counts did not differ significantly between the two morphotypes [Fig. 2(a)], RgT behaved differently towards the uptake of radiolabelled glycerol [Fig. 2(b)] and dimethylthiazol-diphenyltetrazolium bromide (MTT) [Fig. 2(c)]. In spite of an equal number of cfu in both cultures, the RgT MAC failed to take up radiolabelled glycerol and MTT dye, suggesting that structural changes existed in the bacterial envelope. The difference in uptake of MTT was also supported by visible differences in the uptake of this dye by the two types of colonies (Fig. 3). These results prompted us to examine the uptake of other small molecules like alamar blue and neutral red to assess permeability changes in these bacilli.
Colony morphology of MAC affects virulence and biochemical properties
205
1010 (a) 9
10
8
cfu/ml
10
107
106
105
1
2
3
4
5
6
7
8
9
10
60
Figure 3. RgT and SmT MAC colonies (×40) showing the uptake and reduction of MTT.
50 40
0.20 (a)
20 10
0
1
2
3
4
5
6
7
8
9
10
8 (c) Absorbance (570 nm)
7
Absorbance (570–595 nm)
30
0.15
0.10
0.05
6 5
0
2
4 6 Days of culture
8
10
2
4 6 Days of culture
8
10
4
0.3
3
(b)
2 1 0
1
2
3
4 6 5 Days of culture
7
8
9
10
Figure 2. Mycobacterial growth determination of the RgT (Χ) and SmT (Β) morphotypes as assessed by (a) viable counts, (b) radiolabelled glycerol uptake and (c) MTT reduction.
Uptake of alamar blue and neutral red Equal numbers of bacilli obtained from actively growing SmT and RgT MAC cultures were examined for their ability to incorporate alamar
Absorbance (540 nm)
3
H -glycerol uptake (cpm × 1000)
(b)
0.2
0.1
0
Figure 4. Uptake of (a) alamar blue and (b) neutral red by the RgT (Χ) and SmT (Β) colony morphotypes during the growth in MBB.
206
R. G. Kansal et al.
(a)
(b) 1
2
3
1
2
Bst N1
3
4
5
6
7
8
9
Xho1
Figure 5. Gene amplification and restriction fragment length polymorphism analysis: (a) PCR amplification products of SmT (lane 1), RgT (lane 2) and molecular size markers (lane 3). (b) RFLP profiles by using Bst N1 (lanes 1–4) or Xho1 (lanes 5–8) from MAC variants, M. avium ATCC 35713 (lanes 1 and 5), M. intracellulare ATCC 35761 (lanes 2 and 6), RgT (lanes 3 and 7), SmT (lanes 4 and 8) and molecular size markers (lane 9).
blue and neutral red as described in Materials and methods. Interestingly, uptake of alamar blue [Fig. 4(a)] and neutral red [Fig. 4(b)] by RgT bacilli was significantly lower when compared to the SmT bacilli, suggesting that there are major differences in the cell envelopes of the SmT and RgT variants of MAC.
Restriction fragment length polymorphism analysis The 1380 bp amplicon, which is a part of hsp 65 protein and has been shown to be genus specific, could be obtained from all the four strains of mycobacteria, including two ATCC strains, following the PCR reaction. The PCR products from the SmT and RgT MAC are shown in Fig 5(a). The pattern of fragments obtained after digestion of PCR products with the restriction enzymes is shown in Fig. 5(b). The patterns obtained with both RgT and SmT were identical and closer to that of M. avium (characteristic of MAC serotype 1) but were different from M. intracellulare. These findings confirm that RgT and SmT are derived from the same parent strain.
Growth of the SmT and RgT MAC in cultured macrophages The two morphotypes showed significant differences in their ability to infect and multiply in the resident peritoneal macrophages. The number of RgT bacilli phagocytosed by the macrophages (89±9) after 4 h of infection was significantly greater (P<0.001) than that of SmT bacilli (40±9). These numbers increased to 326±29 for RgT and 97±17 for SmT 2 days post-infection. The four-fold increase in cfu of the RgT morphotype, however, does not correlate with the amount of radiolabelled glycerol taken up by the growing bacilli (Table 1) because of the fact that the RgT morphotype does not take up glycerol very well. These results indicate that the RgT variant of the MAC could infect and multiply in macrophages better than the SmT variant.
Virulence of the SmT and RgT MAC in vivo The level of infection produced by the two morphotypes varied significantly as evaluated by the following parameters.
Colony morphology of MAC affects virulence and biochemical properties
207
Table 1. Growth and proliferation of the RgT and SmT MAC in cultured mouse peritoneal macrophages Time postinfection (days) 0a 2
SmT MAC cfu
RgT MAC
cpm
cfu
40±9 ND 97±17 158372±41462
cpm
89±9b ND 326±29b 175714±42834
a Macrophage monolayers were washed of any free bacilli after 4 h of infection, lysed and processed for agar plating as described in Methods (ND; not done). b Significantly (P<0.001) more than SmT.
80
obtained at day 1 post-infection did not vary significantly in all of the tissues. However, as the infection progressed, the RgT MAC proliferated at a much faster rate than the SmT MAC in all of the organs. At day 50 postinfection, the number of bacilli recovered from the lungs, liver and spleen of animals infected with the RgT morphotype was at least two to three log greater (P<0.001) than those recovered from the animals infected with the SmT morphotype [Fig. 7(a–c)].
Percent mortality
60
40
20
0
Blood profile
10
20 30 Days post-infection
40
50
Figure 6. Percent mortality obtained in beige mice infected with the SmT (Β) and RgT (Χ) variants of the MAC. Groups of 20 mice each were injected intravenously with SmT or RgT MAC cultures containing approximately 5×107 bacilli via the tail vein. These mice were observed for mortality up to 40 days post-infection.
Mortality Animals infected with SmT or RgT bacilli showed similar mortality up to 2 weeks postinfection. Although, the mortality in both groups was 10% at day 20, it increased to 50% by day 30 and 70% by day 40 in the group infected with RgT MAC. The mortality in the mice infected with the SmT MAC (10%) remained unchanged throughout the experiment (Fig. 6).
Level of infection in various organs Because the animals were infected with the same number of bacilli (5×107 bacilli/mouse), the cfu
Bacteraemia is a known feature of disseminated MAC infection in patients. The number of bacilli present in blood collected from mice infected with the SmT and RgT MAC is shown in Fig. 7(d). The bacteraemia caused by the SmT and RgT morphotypes could be cleared during an early phase of infection; at day 8 post-infection, no bacilli were detected in the blood. However, in the animals infected with the RgT morphotype, the bacilli reappeared in the blood on day 15 and remained in circulation until day 50 post-infection, whereas no bacteraemia could be detected in mice infected with the SmT MAC after 10 days of infection. The severity of the infection caused by RgT was also supported by other blood parameters of animals infected with this MAC variant. Although the blood profile of the animals infected with SmT appeared to be normal, all parameters such as haemoglobin, haematocrit and the number of red blood cells in blood of the animals infected with RgT were markedly reduced (Table 2), suggesting the presence of a situation similar to anaemia in these mice.
Interconversion of SmT to RgT MAC in various organs of infected mice We observed that the bacilli recovered from
208
R. G. Kansal et al. 10
9
10
10 (a)
(b)
109 108 cfu/gm tissue
cfu/gm tissue
108 7
10
106
107
105 4
10
6
0
10
20
30
40
50
10
60
10
0
10
20
30
40
50
60
10
20 30 40 Days post-infection
50
60
4
10
10 (c)
(d)
109
3
108 cfu/ml
cfu/gm tissue
10
107
101
106
105
102
0
10
20 30 40 Days post-infection
50
60
100
0
Figure 7. Number of bacilli recovered from the lungs (a), liver (b), spleen (c) and blood (d) of beige mice infected with the SmT (Β) and RgT (Χ) morphotypes of the MAC from day 0 to day 50 post-infection. Each time point represents mean±SD of values obtained from three to four animals.
Table 2. Blood profile of mice infected with the SmT and RgT MAC Blood parameter RBC×106 Haemoglobin (g/dl) Haematocrit
SmT
RgT
9.99±0.13 14.3±0.2 38.7±0.52
5.73±0.87 7.4±1.5 23.3±3.9
Blood was collected on day 35 post-infection.
animals infected with the RgT colony morphotype did not show any change in colony morphology, whereas those infected with the SmT MAC were gradually converted to the RgT
morphotype. The number of RgT variants appearing on agar plate cultures obtained from mice infected with SmT bacilli were counted and the conversion percentage was calculated. As shown in Table 3, the percentage of conversion to RgT increased along with the disease progression. By day 35 post-infection, all of the recovered bacterial colonies were converted to the RgT variant.
Maintenance of the RgT variant in macrophages The continuous growth of the RgT variant in broth causes frequent conversions to the SmT
Colony morphology of MAC affects virulence and biochemical properties
209
Table 3. Interconversion of the SmT and RgT morphotypes in organs of the infected mice Infecting Day postmorphotype infection
Sm-T
1 8 20 35 50
Conversion to the other morphotype (%) Lung
Liver
Spleen
Nil 4.2±1.9 100 100 100
Nil 2.2±0.4 89 ±3.6 100 100
Nil 5.1±2.1 91.3±3.2 100 100
The bacilli recovered from animals infected with RgT colony morphotype did not show any change in colony morphology.
variant, which possesses low virulence. When grown in cultured macrophages, both the colony morphology and the virulence of RgT can be maintained. We have examined the morphotype of the MAC recovered from macrophages infected with the RgT variant for five cycles of 1 week each. During this period, the RgT colony morphology was maintained and did not revert back to SmT, suggesting that the interaction of the RgT variant with macrophages stabilized its morphology and virulence as well as the other characteristics discussed above.
Discussion MAC isolates are known to form multiple types of colonies on solid media. These different colony types were initially used for bacterial identification [14]. Some studies, however, reported the relationship of colony morphology to differences in virulence [10]. All of these studies were inconclusive because of the high rates of interconversion of one form to the other and thus hampered a more systematic study of the virulence factors. Our present data show a direct correlation of the RgT colony morphology to metabolic changes and increased virulence of the MAC. These studies will be helpful in identifying the stable biochemical markers for MAC virulence. Our initial observations were the differences in the turbidity of the SmT and RgT cultures, the morphology of the colonies, and glycerol uptake by the SmT and RgT MAC which prompted further investigations. Similar growth rates of the two morphotypes ruled out the possibility of
differences in generation time [Fig. 2(a)]. Differences in the uptake of glycerol and in MTT reduction [Fig. 2(b) and (c)], however, suggested that the two variants may differ in structural components and/or metabolic activity. Similarly, differences in the uptake of other different-sized molecules used as dyes (alamar blue and neutral red; Fig. 4) indicated possible major changes in the cell envelopes of these MAC variants. The two colony morphotypes differed significantly in their ability to infect and proliferate in susceptible hosts such as the beige mice used in this study. Although the BALB/c mouse strain has also been used [15], the beige mouse model is more widely accepted for studying MAC infections [13, 16–18]. The number of bacteria recovered from various organs has been shown to be a reliable measure of virulence and has been used for studying the virulence of different MAC isolates obtained from patients [19, 20]. Our results show that the RgT MAC proliferates in vivo at a much faster rate than the SmT MAC, resulting in higher mortality for infected animals. This could be caused by the higher resistance of the RgT MAC to the antimicrobial arsenal of macrophages than that of SmT MAC as determined by assaying macrophage microbicidal activity. As evidenced in macrophage cultures infected with the same number of SmT and RgT bacilli, the number of RgT MAC recovered from cultured macrophages 2 days postinfection was four times higher than that of SmT MAC (Table 1). There are controversial reports regarding the in vitro and in vivo correlation of virulence. Cultured human monocytes have been used by some to analyse the virulence of mycobacteria [21, 22] but others believe that it might give an
210
erroneous estimate of mycobacterial virulence [15]. Our study shows that varied results could be obtained by using different methods such as the discrepancy in results obtained by the two methods used to assess MAC growth in macrophages (cfu counting and radiolabelled glycerol uptake). We believe that the structural changes in bacilli could lead to differences in metabolic activity, ultimately affecting the outcome of the study. Therefore, our data emphasize the reliability of cfu counts over other methods, particularly when morphological and structural changes in bacilli are involved. Various authors in the past have reported differences in the virulence of various colony morphotypes, but these studies provided epidemiological information rather than models for studying the virulence factors associated with these bacteria. Other studies have used standard strains known to vary in virulence for studying interactions with the cultured macrophages [23, 24]. The major obstacle in conducting a more systematic study was the high frequency of interconversion of these colony forms. We have observed that the SmT MAC could be converted to the RgT MAC progressively with the time of infection in beige mice (Table 3). However, once the RgT variant was obtained, it seldom reverted back to SmT. In addition, the colony morphology and hence the virulence of RgT bacilli could be maintained indefinitely by continuous passage in macrophages. Thus, we have obtained two distinct colony morphotypes of the MAC that originated from the same culture but possess major differences in their morphology, metabolic activity and virulence. Our data indicated significant changes in either the permeability or certain metabolic pathways of the MAC during conversion from the SmT MAC to the RgT MAC, leading to reduced incorporation of compounds such as glycerol, MTT, alamar blue and neutral red. The failure of the RgT MAC to reduce MTT, alamar blue and neutral red might also indicate the possibility of a deficiency in some enzyme systems necessary for the incorporation and metabolization of these compounds. Further studies to delineate the differences in structural or biochemical components between these variants are in progress.
Materials and methods MAC isolate The MAC 101 strain, obtained from Childrens’ Hospital, Los Angeles, CA, U.S.A., was main-
R. G. Kansal et al.
tained on slants of Lowenstein-Jensen medium (Remel, Lenexa, KS, U.S.A.). Culture suspensions were prepared by growing an initial inoculum on Middlebrook 7H11 agar (Difco Laboratories, Detroit, MI, U.S.A.) supplemented with OADC (oleate, albumin, dextrose and catalase) enrichment (Remel) and glycerol (Boehringer Mannheim Biochemicals, Indianapolis, IN, U.S.A.) as described previously [16]. Both types of colonies from these plates were subcultured in Middlebrook 7H9 broth (MBB) (Difco Laboratories) supplemented with OADC enrichment and glycerol and allowed to grow at 37°C on a rotary shaker for 3 days. The culture suspensions were shaken and sonicated for 90 s using a bath-type sonicator (Laboratory Supplies Co., Inc., Hicksville, NY, U.S.A.) and adjusted to the desired concentrations using a Klett Summerson colorimeter (Klett Manufacturing, Brooklyn, NY, U.S.A.) or a spectronic-20 spectrophotometer (Milton Roy, Rochester, NY, U.S.A.). The size of the inoculum was determined by titration, plating and counting CFU on 7H11 agar plates after 2 weeks of incubation at 37°C. The cultures were diluted appropriately to contain equal numbers of bacilli in both SmT and RgT MAC suspensions and kept at −20°C until use.
Animals Female C57 BL/6 J/bgj bgj mice aged 4–6 weeks (weight, 14–16 g) were obtained from Jackson Laboratories (Bar Harbor, ME, U.S.A.). These animals received water and food ad libitum. Animals were used according to a protocol approved by the Institutional Animal Care and Use Committee at the University of Texas M. D. Anderson Cancer Center.
Determination of mycobacterial growth The growth rates of the SmT MAC and the RgT MAC were determined by counting cfu at different times during the MAC culture in broth. Briefly, 10 ml of MBB were inoculated with a known number of bacilli from each of the colony forms and allowed to grow at 37°C on a rotary shaker (160 rpm). Aliquots were taken out at days 0, 1, 2, 4, 6 and 8, and the cfu were determined by plating appropriate dilutions on 7H11 agar plates. The colonies were counted after 2 weeks of incubation at 37°C.
Colony morphology of MAC affects virulence and biochemical properties
[3H]-glycerol uptake assay Both the SmT and RgT MAC were grown in MBB without glycerol but containing [3H]-glycerol (ICN Biochemicals, Irvine, CA, U.S.A.) at a concentration of 50 lCi/ml under shaking conditions at 37°C. One hundred ll aliquots of both MAC types were taken out in microtitre wells on days 0, 1, 2, 4, 6 and 8 of the growth. Fifty microlitres of 5.25% sodium hypochlorite were added to each well, and bacteria were harvested onto fibreglass filters with a semiautomatic cell harvester (Cambridge Technology, Watertown, MA, U.S.A.). The filter-paper discs were dried overnight in separate vials and analysed for radioactivity incorporated into the bacteria by using Scintiverse II (Fisher Scientific, Fair Lawn, NJ, U.S.A.) and a scintillation counter as described previously [17].
MTT reduction assay The MTT reduction assay was done according to a colorimetric method developed previously in our laboratory [25]. Briefly, 10 ml of MBB were inoculated with a known equal number of bacilli from both the SmT or RgT MAC cultures and allowed to grow at 37°C on a rotary shaker (160 rpm). One hundred microlitre aliquots of both types of bacilli were taken out in microtiter plates on days 0, 1, 2, 4, 6 and 8 of growth. Ten microlitres of the 5 mg/ml stock solution of MTT were added to these aliquots and incubated for 3 h at 37°C. At the end of the incubation, 100 ll of the extraction buffer was added to all the wells. After overnight incubation at 37°C, absorbance was measured at 570 nm with a microplate reader (Molecular Devices, Menlo Park, CA, U.S.A.).
Alamar blue assay Alamar blue (Alamar, Sacramento, CA, U.S.A.) is an indicator dye, formulated to measure the proliferation of a variety of eukaryotic cells as well as bacteria quantitatively. A 10% solution of alamar blue in MBB was prepared. A known equal number of both SmT- or RgT-morphotype bacilli were inoculated separately in tubes containing 10 ml of the above solution and incubated at 37°C on a rotary shaker (160 rpm). 100 ll aliquots (in triplicate) were taken out in wells of the microtitre plates on days 0, 1, 2, 4,
211
6 and 8, and the absorbance was read at 570 and 595 nm. The difference in absorbance between 570 and 595 nm was taken as the overall absorbance value.
Neutral red uptake 10 ml of MBB were inoculated with a known, equal number of bacilli from both the SmT or RgT MAC cultures and incubated at 37°C on a rotary shaker (160 rpm). Four hundred microlitre aliquots of each culture were taken out in microfuge tubes on days 0, 1, 2, 4, 6 and 8 of growth and mixed with 8 ll of 0.1% neutral red solution. The mixture was vortexed and incubated for 1 h at 37°C. The cells were then centrifuged at 15 000 g for 15 min and washed twice with phosphate-buffered saline (PBS). 300 ll of the developer (ethanol:acetic acid: water; in v/v ratio 368:3:129) were added to each tube; the contents were mixed, transferred to a 96-well plate, and read at 540 nm.
Restriction fragment length polymorphism analysis To ascertain that the RgT isolate was a variant of the parent strain and not a contaminant, a two-step assay combining a gene amplification step and a restriction fragment length polymorphism analysis, developed previously by Plikaytis et al. [26] was used.
Sample preparation The genomic DNA to be used for amplification was prepared from RgT, SmT as well as reference strains of M. avium (ATCC 35713) and M. intracellulare (ATCC 35761) as described previously [27]. Briefly, the bacterial cell suspensions were prepared in Tris EDTA (TE) buffer and were heat inactivated for 10 min at 80°C, followed by centrifugation at 13 000 g at 4°C for 15 min. The supernatants were then used as source of template DNA for PCR.
Gene amplification Primers corresponding to portions of the M. tuberculosis hsp 65 gene were obtained from
212
GIBCO BRL custom primers of Life Technologies, Gaithersburg, MD, U.S.A. and used to obtain an amplicon on a thermal cycler as described earlier [27].
Restriction enzyme digestion Fifteen microlitres of the amplified reaction mixture was transferred to a fresh tube and digested with either Bst NI (New England Biolabs, Inc., Beverly, MA, U.S.A.) or Xhol (Bethesda Research Laboratories, Gaithersburg, MD, U.S.A.). The products obtained were analysed on 1.5% agarose gels and the DNA bands were visualized by ethidium bromide fluorescence. Gels were photographed with Polaroid type 55 positive/ negative film (Polaroid Corp., Cambridge, MA, U.S.A.).
Growth of the SmT and RgT MAC in cultured macrophages: macrophage harvest and culture Mice were killed by cervical dislocation, and peritoneal macrophages were harvested in cold RPMI 1640 (GIBCO, Gaithersburg, MD, U.S.A.). The cell suspension was centrifuged at 4°C, 200 g for 10 min and resuspended in prewarmed RPMI containing 10% fetal bovine serum, 1% penicillin-streptomycin solution and 1% HEPES. The cell number was determined with a coulter counter (Coulter Electronics, Hialeah, FL, U.S.A.). Cells were then seeded into 96-well plates (Becton Dickinson Co., Lincoln Park, NJ, U.S.A.) at a concentration of 1.5×105 cells/well. After incubation for 2 h at 37°C in 5% CO2, nonadherent cells were removed by washing twice with RPMI 1640 as described elsewhere [17]. Monolayers were then incubated overnight at 37°C in a humidified 5% CO2 incubator.
R. G. Kansal et al.
bacilli and reconstituted with 200 ll of fresh medium. To determine the number of bacilli taken up by the macrophages after 4 h of phagocytosis, half of the wells were lysed at this time and processed for the determination of cfu. The rest of the infected macrophages were incubated further to let the bacilli grow and multiply. After 2 days of incubation, these macrophages were lysed and processed as described below.
Assay of intracellular MAC growth The growth of MAC in macrophages was studied by measuring [3H]-glycerol uptake by viable bacteria or by determining the number of cfu in the macrophage lysate as described previously [17]. Briefly, the infected cells were lysed at the end of the incubation period by removing the medium and replacing it with 100 ll MBB (without added glycerol) followed by the addition of 10 ll 1 N NaOH, 10 ll 1 N HCl (to neutralize NaOH), and 10 ll 5% Tween-80 (Difco Laboratories). For determination of CFU, appropriate dilutions of the macrophage lysate were prepared in PBS, and 100-ll aliquots were plated onto 7H11 agar. Plates were incubated for 12–14 days until the colonies were apparent and cfu were counted. To measure the incorporation of radioactivity by viable bacilli, 1 lCi/well of [3H]-glycerol was added to all the wells and the plates were incubated for 4 days at 37°C. Next, 50 ll of 5.25% sodium hypochlorite were added to all of the wells and bacteria were harvested onto fibreglass filters using a semiautomatic cell harvester. The filter-paper discs were dried overnight in separate vials and analysed for radioactivity incorporated into the bacteria using scintiverse II and a scintillation counter (Packard Instrument Company, Laguna Hills, CA, U.S.A.).
Infection of macrophage cultures
Evaluation of virulence in vivo
After overnight culture, monolayers were washed twice with Excell 300 (JRH Biosciences, Lenexa, KS, U.S.A.) containing 1% HEPES. Monolayers were then infected with 20 ll of culture suspensions containing 1×107 cfus of the SmT and RgT MAC (MAC-to-macrophage ratio, 10:1) for 4 h at 37°C in 5% CO2 [17]. After incubation, infected cells were washed three times with Excell 300 to remove extracellular
The virulence of the SmT and RgT MAC was studied in the beige mice by comparing the mortality of the mice and cfu recovered from the lungs, liver and spleen in the respective groups at various time intervals post-infection as described previously [16]. Groups of 20 mice each received intravenous injections of SmT or RgT MAC cultures containing approximately 5×107 bacilli through the tail vein. These mice
Colony morphology of MAC affects virulence and biochemical properties
213
were observed for mortality up to 60 days postinfection, and the results obtained from mice infected with the RgT MAC were compared with those obtained from mice infected with SmT MAC.
Coordinating Board to RTM and by the National Institute of Health-National Cancer Institute (core) support grant CA-16672 to the Department of Veterinary Medicine and Surgery, The University of Texas M. D. Anderson Cancer Center. We thank Don Norwood for editorial assistance.
Assessment of infection in various organs
References
Three to four animals from both groups were killed on days 1, 8, 15 and 22 post-infection, and vital organs such as the lungs, liver and spleen were removed aseptically. The tissues were weighed and homogenized in a known volume of PBS using an electric homogenizer (Omni International, Inc., Gainesville, VA, U.S.A.). The cfu were determined after plating the appropriate dilutions on 7H11 agar. The plates were incubated at 37°C, and cfu were counted after 2 weeks of incubation. The results were expressed as cfu/g of tissue. To examine the conversion of the SmT morphotype to RgT morphotype, both colony types from each organ were counted and the conversion percentage was calculated.
Bacteraemia and blood profile Blood was collected post-infection from the retro-orbital plexus of the infected animals at the indicated time periods, and a 100-ll aliquot was plated on the MBB agar plates to find out the extent of bacteraemia in animals infected with SmT and RgT bacilli. Other blood parameters such has haemoglobin, haematocrit and the number of red blood cells were also examined on day 35 post-infection when the infection was well established.
Statistical analysis The results were expressed as the mean ±SD of triplicates from one representative experiment; all experiments were performed three times with similar results. The level of significance was assessed by Student’s t-test.
Acknowledgements This work was supported by grants ATP-D 000015084 and ATP 000015091 from the Texas Higher Education
1 Horsburgh CR Jr., Selik RM. The epidemiology of disseminated nontuberculous mycobacterial infection in the acquired immunodeficiency syndrome (AIDS). Am Rev Resp Dis 1989; 139: 4–7. 2 Macher AM, Kovacs JA, Gill V et al. Bacteremia due to Mycobacterium avium-intracellulare in the acquired immunodificiency syndrome. Ann Intern Med 1983; 99: 782–5. 3 Murray JF, Felton CP, Garay SM et al. Pulmonary complications of the acquired immunodeficiency syndrome. N Engl J Med 1984; 310: 1682–8. 4 Young LS. Mycobacterium avium complex infection. J Infect Dis 1988; 157: 863–7. 5 Pattyn SR, Hermans-Boveroulle MT. Dissociation in M. avium. Pneumonology 1970; 142: 119–25. 6 Inderlied CB, Kemper CA, Bermudez LE. The Mycobacterium avium complex. Clin Microbiol Rev 1993; 6: 266–310. 7 Benson CA. Treatment of disseminated disease due to M. avium complex in patients with AIDS. Clin Infect Dis 1994; 18(suppl 3): S237–42. 8 Pattyn SR. A study of group 111 non chromogenic mycobacteria. Correlation of chicken virulence with other in vitro characters among 20 strains. Zeitsch fu¨r Tuberk 1967; 127: 41–6. 9 Moehring JM, Solotorovsky MR. Relationship of colonial morphology to virulence for chickens of Mycobacterium avium and the non-photochromogens. American Review of Respiratory Disease 1965; 92: 704–13. 10 Schaefer WB, Davis CL, Cohn ML. Pathogenicity of transparent, opaque and rough variants of Mycobacterium avium in chickens and mice. Am Rev Resp Dis 1970; 102: 499–506. 11 Anz W, Meissner G. Comparative virulence test in chickens with transparent and opaque strains of avian mycobacteria of different serotypes. Zentralblatt Fur Bakteriologie, Parasitenkunde, Infektionskranheiten Und Hygiene 1972; 221: 334–42. 12 Reddy VM, Luna-Herrera J, Gangadharam PRJ. Pathobiological significance of colony morphology in Mycobacterium avium complex. Microb Pathog 1996; 21: 97–109. 13 Gangadharam PRJ. Beige mouse model for Mycobacterium avium complex disease. Antimicrob Agents Chemother 1995; 39: 1647–54. 14 Kuze F, Uchihira F. Various colony formers of Mycobacterium avium-intracellulare. Eur J Resp Dis 1984; 65: 402–10. 15 Pedrosa J, Florido M, Kunze ZM et al. Characterization of the virulence of Mycobacterium avium Complex (MAC) isolates in mice. Clin Exp Immunol 1994; 98: 210–16. 16 Mehta RT. Liposome encapsulation of clofazimine reduces toxicity in vitro and in vivo and improves therapeutic efficacy in the beige mouse model of
214
17
18
19
20
21
disseminated Mycobacterium avium-M. intracellulare complex infection. Antimicrob Agents Chemother 1996; 40: 1893–902. Gomez-Flores R, Tucker SD, Kansal RG, Tamez-Guerra R, Mehta RT. Enhancement of antibacterial activity of clofazimine against Mycobacterium avium-intracellulare complex infection induced by IFN-c is mediated by TNF-a. J Antimicrob Chemother 1997; 39: 189–97. Kansal RG, Gomez-Flores R, Sinha I, Mehta RT. Therapeutic efficacy of liposomal clofazimine against Mycobacterium avium complex in mice depends on size of initial inoculum and duration of infection. Antimicrob Agents Chemother 1997; 41: 17–23. Gangadharam PRJ, Perumal VK, Jairam BT, Podapati R, Taylor RB, LeBrecque JF. Virulence of Mycobacterium avium complex strains from acquired immune deficiency syndrome patients: relationship with characteristics of the parasite and host. Microb Pathog 1989; 7: 263–78. Reddy VM, Parikh K, Luna-Herrera J, Falkinham JO, Brown S, Gangadharam PRJ. Comparison of virulence of Mycobacterium avium complex (MAC) strains isolated from AIDS and non-AIDS patients. Microb Pathog 1994; 16: 121–30. Toba H, Crawford JT, Ellner JJ. Pathogenicity of Mycobacterium avium for human monocytes: absence of macrophage activating factor activity of gamma interferon. Infect Immun 1989; 57: 239–44.
R. G. Kansal et al. 22 Crowle AJ, Tsang AY, Vatter AE, May MH. Comparison of 15 laboratory and patient-derived strains of Mycobacterium avium for ability to infect and multiply in cultured human macrophages. J Clin Microbiol 1986; 24: 812–21. 23 Sarmento AM, Appelberg R. Relationship between virulence of Mycobacterium avium strains and induction of tumor necrosis factor alpha production in infected mice and in in vitro-cultured mouse macrophages. Infect Immun 1995; 63: 3759–64. 24 Sarmento AM, Appelberg R. Involvement of reactive oxygen intermediates in tumor necrosis factor alphadependent bacteriostasis of Mycobacterium avium. Infect Immun 1996; 64: 3224–30. 25 Gomez-Flores R, Gupta S, Tamez-Guerra R, Mehta RT. Determination of MICs for Mycobacterium avium-M. intracellulare complex in liquid medium by a colorimetric method. J Clin Microbiol 1995; 33: 1842–6. 26 Plikaytis BB, Plikaytis BD, Yakrus MA et al. Differentiation of slowly growing Mycobacteria species, including Mycobacterium tuberculosis, be gene amplification and restriction fragment length polymorphism analysis. J Clin Microbiol 1992; 30: 1815–22. 27 Telenti A, Marchesi F, Balz M, Bally F, Bottger EC, Bodmer T. Rapid identification of Mycobacteria to the species level by polymerase chain reaction and restriction enzyme analysis. J Clin Microbiol 1993; 31: 175–8.
MICROBIAL PATHOGENESIS
Change in colony morphology influences the virulence as well as the biochemical properties of the Mycobacterium avium complex Rita G. Kansal, Ricardo Gomez-Flores & Reeta T. Mehta∗ Department of Bioimmunotherapy, The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030 U.S.A. (Received October 7, 1997; accepted in revised form August 11, 1998)
Factors that influence colony morphology are of crucial importance for drug development as well as for understanding the virulence of Mycobacterium avium complex (MAC) strains. The MAC 101 strain used in the present study grows as smooth transparent (SmT) colonies that tend to become opaque and pigmented when incubated for long periods of time. However, when MAC was passaged in animals, two types of colonies were recovered. The new rough transparent (RgT) colony morphology appeared more flat and transparent, having a central spot, irregular edges at times, and a dry, granular appearance like that of the rough mutants. In animal studies, the RgT bacilli multiplied at a much faster rate than that of the SmT bacilli, causing 60–80% mortality compared with the 10% mortality observed in mice infected with SmT. In vitro studies indicated that the SmT MAC did not grow and multiply as well in resident peritoneal macrophages as the RgT MAC did. The two morphotypes did not differ in their growth rates in vitro but the RgT MAC failed to reduce dimethylthiazol-diphenyltetrazolium bromide (MTT), alamar blue and neutral red, suggesting that there might be significant changes in the cell wall or elsewhere causing changes in cellular permeability. These two morphotypes could serve as models for studying the biochemical markers or the identification of factors responsible for the virulence of the MAC. 1998 Academic Press
Key words: Mycobacterium avium complex, colony morphology, virulence, growth rate, biochemical properties, MTT reduction.
Introduction ∗ Author for correspondence at: Department of Bioimmunotherapy, Box 60, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030, U.S.A. 0882–4010/98/100203+12 $30.00/0
The Mycobacterium avium complex (MAC) belongs to a group of bacteria known as atypical mycobacteria. Infection caused by the MAC is one of the most common opportunistic infections 1998 Academic Press
204
afflicting patients with advanced acquired immunodeficiency syndrome (AIDS) [1–4]. There is compelling evidence that this infection in humans contributes to a decrease in both quality of life and survival [5, 6]; its effective treatment does prolong survival as well as quality of life [7]. The pathogenesis of the MAC is not well understood. However, the colony morphology of MAC, distinguished into three groups (SmT, SmOp and RgT) has been associated with virulence in chickens and mice [8–11]. Although factors that influence colony morphology are of crucial importance, existing studies do not provide conclusive information; one of the important factors contributing to this is the high rate of interconversions of these colonies [12]. It has been reported that the virulence of the infecting strain of MAC 101 increases markedly following passage in animals [13], but no comparison with colony morphology is available. The MAC 101 strain used in the present study grows as smooth transparent (SmT) colonies that tend to become opaque and pigmented after prolonged incubation. However, when passaged in animals, a new colony morphotype emerged that appeared flatter with a central spot, irregular edges at times and a dry granular appearance like that of the rough transparent (RgT) mutants. The bacilli recovered from infected organs formed RgT type colonies that increased in number as the time of infection increased. This morphotype could be maintained in macrophages without passage in the animals. In an attempt to study the underlying mechanisms that control colony morphology, the two morphotypes were isolated and studied with respect to their growth, metabolic status, and virulence.
Results Morphological appearance MAC 101 grows as SmT colonies on 7H11 agar that tend to become opaque and pigmented after prolonged incubation. However, when passaged in animals, two colony types could be recovered. The new colony morphotype (RgT) appeared flatter with irregular edges at times and was dry in appearance like the rough mutants. These colonies look similar and are difficult to distinguish until 10 days after agar plating of the culture; as they grow older, the SmT colonies
R. G. Kansal et al.
Figure 1. Typical colony morphology of 21-day-old SmT and RgT colonies under simple microscopy (×40).
become more opaque and pigmented than the RgT colonies which remain transparent and white. Fig. 1 shows the typical colony morphology of 21 day old SmT and RgT colonies under light microscopy (×100). These two morphotypes were then isolated and grown separately in Middlebrook 7H9 broth (MBB) for further studies. Broth cultures of the SmT and RgT MAC showed visible differences in turbidity (the RgT MAC had much less turbidity than the SmT MAC) with no differences in the number of colony forming units (cfu) per ml [Fig. 2(a)].
Mycobacterial growth determination The growth of the SmT and RgT morphotypes in MBB from day 1 to day 8 yielded interesting results. Whereas growth in terms of viable counts did not differ significantly between the two morphotypes [Fig. 2(a)], RgT behaved differently towards the uptake of radiolabelled glycerol [Fig. 2(b)] and dimethylthiazol-diphenyltetrazolium bromide (MTT) [Fig. 2(c)]. In spite of an equal number of cfu in both cultures, the RgT MAC failed to take up radiolabelled glycerol and MTT dye, suggesting that structural changes existed in the bacterial envelope. The difference in uptake of MTT was also supported by visible differences in the uptake of this dye by the two types of colonies (Fig. 3). These results prompted us to examine the uptake of other small molecules like alamar blue and neutral red to assess permeability changes in these bacilli.
Colony morphology of MAC affects virulence and biochemical properties
205
1010 (a) 9
10
8
cfu/ml
10
107
106
105
1
2
3
4
5
6
7
8
9
10
60
Figure 3. RgT and SmT MAC colonies (×40) showing the uptake and reduction of MTT.
50 40
0.20 (a)
20 10
0
1
2
3
4
5
6
7
8
9
10
8 (c) Absorbance (570 nm)
7
Absorbance (570–595 nm)
30
0.15
0.10
0.05
6 5
0
2
4 6 Days of culture
8
10
2
4 6 Days of culture
8
10
4
0.3
3
(b)
2 1 0
1
2
3
4 6 5 Days of culture
7
8
9
10
Figure 2. Mycobacterial growth determination of the RgT (Χ) and SmT (Β) morphotypes as assessed by (a) viable counts, (b) radiolabelled glycerol uptake and (c) MTT reduction.
Uptake of alamar blue and neutral red Equal numbers of bacilli obtained from actively growing SmT and RgT MAC cultures were examined for their ability to incorporate alamar
Absorbance (540 nm)
3
H -glycerol uptake (cpm × 1000)
(b)
0.2
0.1
0
Figure 4. Uptake of (a) alamar blue and (b) neutral red by the RgT (Χ) and SmT (Β) colony morphotypes during the growth in MBB.
206
R. G. Kansal et al.
(a)
(b) 1
2
3
1
2
Bst N1
3
4
5
6
7
8
9
Xho1
Figure 5. Gene amplification and restriction fragment length polymorphism analysis: (a) PCR amplification products of SmT (lane 1), RgT (lane 2) and molecular size markers (lane 3). (b) RFLP profiles by using Bst N1 (lanes 1–4) or Xho1 (lanes 5–8) from MAC variants, M. avium ATCC 35713 (lanes 1 and 5), M. intracellulare ATCC 35761 (lanes 2 and 6), RgT (lanes 3 and 7), SmT (lanes 4 and 8) and molecular size markers (lane 9).
blue and neutral red as described in Materials and methods. Interestingly, uptake of alamar blue [Fig. 4(a)] and neutral red [Fig. 4(b)] by RgT bacilli was significantly lower when compared to the SmT bacilli, suggesting that there are major differences in the cell envelopes of the SmT and RgT variants of MAC.
Restriction fragment length polymorphism analysis The 1380 bp amplicon, which is a part of hsp 65 protein and has been shown to be genus specific, could be obtained from all the four strains of mycobacteria, including two ATCC strains, following the PCR reaction. The PCR products from the SmT and RgT MAC are shown in Fig 5(a). The pattern of fragments obtained after digestion of PCR products with the restriction enzymes is shown in Fig. 5(b). The patterns obtained with both RgT and SmT were identical and closer to that of M. avium (characteristic of MAC serotype 1) but were different from M. intracellulare. These findings confirm that RgT and SmT are derived from the same parent strain.
Growth of the SmT and RgT MAC in cultured macrophages The two morphotypes showed significant differences in their ability to infect and multiply in the resident peritoneal macrophages. The number of RgT bacilli phagocytosed by the macrophages (89±9) after 4 h of infection was significantly greater (P<0.001) than that of SmT bacilli (40±9). These numbers increased to 326±29 for RgT and 97±17 for SmT 2 days post-infection. The four-fold increase in cfu of the RgT morphotype, however, does not correlate with the amount of radiolabelled glycerol taken up by the growing bacilli (Table 1) because of the fact that the RgT morphotype does not take up glycerol very well. These results indicate that the RgT variant of the MAC could infect and multiply in macrophages better than the SmT variant.
Virulence of the SmT and RgT MAC in vivo The level of infection produced by the two morphotypes varied significantly as evaluated by the following parameters.
Colony morphology of MAC affects virulence and biochemical properties
207
Table 1. Growth and proliferation of the RgT and SmT MAC in cultured mouse peritoneal macrophages Time postinfection (days) 0a 2
SmT MAC cfu
RgT MAC
cpm
cfu
40±9 ND 97±17 158372±41462
cpm
89±9b ND 326±29b 175714±42834
a Macrophage monolayers were washed of any free bacilli after 4 h of infection, lysed and processed for agar plating as described in Methods (ND; not done). b Significantly (P<0.001) more than SmT.
80
obtained at day 1 post-infection did not vary significantly in all of the tissues. However, as the infection progressed, the RgT MAC proliferated at a much faster rate than the SmT MAC in all of the organs. At day 50 postinfection, the number of bacilli recovered from the lungs, liver and spleen of animals infected with the RgT morphotype was at least two to three log greater (P<0.001) than those recovered from the animals infected with the SmT morphotype [Fig. 7(a–c)].
Percent mortality
60
40
20
0
Blood profile
10
20 30 Days post-infection
40
50
Figure 6. Percent mortality obtained in beige mice infected with the SmT (Β) and RgT (Χ) variants of the MAC. Groups of 20 mice each were injected intravenously with SmT or RgT MAC cultures containing approximately 5×107 bacilli via the tail vein. These mice were observed for mortality up to 40 days post-infection.
Mortality Animals infected with SmT or RgT bacilli showed similar mortality up to 2 weeks postinfection. Although, the mortality in both groups was 10% at day 20, it increased to 50% by day 30 and 70% by day 40 in the group infected with RgT MAC. The mortality in the mice infected with the SmT MAC (10%) remained unchanged throughout the experiment (Fig. 6).
Level of infection in various organs Because the animals were infected with the same number of bacilli (5×107 bacilli/mouse), the cfu
Bacteraemia is a known feature of disseminated MAC infection in patients. The number of bacilli present in blood collected from mice infected with the SmT and RgT MAC is shown in Fig. 7(d). The bacteraemia caused by the SmT and RgT morphotypes could be cleared during an early phase of infection; at day 8 post-infection, no bacilli were detected in the blood. However, in the animals infected with the RgT morphotype, the bacilli reappeared in the blood on day 15 and remained in circulation until day 50 post-infection, whereas no bacteraemia could be detected in mice infected with the SmT MAC after 10 days of infection. The severity of the infection caused by RgT was also supported by other blood parameters of animals infected with this MAC variant. Although the blood profile of the animals infected with SmT appeared to be normal, all parameters such as haemoglobin, haematocrit and the number of red blood cells in blood of the animals infected with RgT were markedly reduced (Table 2), suggesting the presence of a situation similar to anaemia in these mice.
Interconversion of SmT to RgT MAC in various organs of infected mice We observed that the bacilli recovered from
208
R. G. Kansal et al. 10
9
10
10 (a)
(b)
109 108 cfu/gm tissue
cfu/gm tissue
108 7
10
106
107
105 4
10
6
0
10
20
30
40
50
10
60
10
0
10
20
30
40
50
60
10
20 30 40 Days post-infection
50
60
4
10
10 (c)
(d)
109
3
108 cfu/ml
cfu/gm tissue
10
107
101
106
105
102
0
10
20 30 40 Days post-infection
50
60
100
0
Figure 7. Number of bacilli recovered from the lungs (a), liver (b), spleen (c) and blood (d) of beige mice infected with the SmT (Β) and RgT (Χ) morphotypes of the MAC from day 0 to day 50 post-infection. Each time point represents mean±SD of values obtained from three to four animals.
Table 2. Blood profile of mice infected with the SmT and RgT MAC Blood parameter RBC×106 Haemoglobin (g/dl) Haematocrit
SmT
RgT
9.99±0.13 14.3±0.2 38.7±0.52
5.73±0.87 7.4±1.5 23.3±3.9
Blood was collected on day 35 post-infection.
animals infected with the RgT colony morphotype did not show any change in colony morphology, whereas those infected with the SmT MAC were gradually converted to the RgT
morphotype. The number of RgT variants appearing on agar plate cultures obtained from mice infected with SmT bacilli were counted and the conversion percentage was calculated. As shown in Table 3, the percentage of conversion to RgT increased along with the disease progression. By day 35 post-infection, all of the recovered bacterial colonies were converted to the RgT variant.
Maintenance of the RgT variant in macrophages The continuous growth of the RgT variant in broth causes frequent conversions to the SmT
Colony morphology of MAC affects virulence and biochemical properties
209
Table 3. Interconversion of the SmT and RgT morphotypes in organs of the infected mice Infecting Day postmorphotype infection
Sm-T
1 8 20 35 50
Conversion to the other morphotype (%) Lung
Liver
Spleen
Nil 4.2±1.9 100 100 100
Nil 2.2±0.4 89 ±3.6 100 100
Nil 5.1±2.1 91.3±3.2 100 100
The bacilli recovered from animals infected with RgT colony morphotype did not show any change in colony morphology.
variant, which possesses low virulence. When grown in cultured macrophages, both the colony morphology and the virulence of RgT can be maintained. We have examined the morphotype of the MAC recovered from macrophages infected with the RgT variant for five cycles of 1 week each. During this period, the RgT colony morphology was maintained and did not revert back to SmT, suggesting that the interaction of the RgT variant with macrophages stabilized its morphology and virulence as well as the other characteristics discussed above.
Discussion MAC isolates are known to form multiple types of colonies on solid media. These different colony types were initially used for bacterial identification [14]. Some studies, however, reported the relationship of colony morphology to differences in virulence [10]. All of these studies were inconclusive because of the high rates of interconversion of one form to the other and thus hampered a more systematic study of the virulence factors. Our present data show a direct correlation of the RgT colony morphology to metabolic changes and increased virulence of the MAC. These studies will be helpful in identifying the stable biochemical markers for MAC virulence. Our initial observations were the differences in the turbidity of the SmT and RgT cultures, the morphology of the colonies, and glycerol uptake by the SmT and RgT MAC which prompted further investigations. Similar growth rates of the two morphotypes ruled out the possibility of
differences in generation time [Fig. 2(a)]. Differences in the uptake of glycerol and in MTT reduction [Fig. 2(b) and (c)], however, suggested that the two variants may differ in structural components and/or metabolic activity. Similarly, differences in the uptake of other different-sized molecules used as dyes (alamar blue and neutral red; Fig. 4) indicated possible major changes in the cell envelopes of these MAC variants. The two colony morphotypes differed significantly in their ability to infect and proliferate in susceptible hosts such as the beige mice used in this study. Although the BALB/c mouse strain has also been used [15], the beige mouse model is more widely accepted for studying MAC infections [13, 16–18]. The number of bacteria recovered from various organs has been shown to be a reliable measure of virulence and has been used for studying the virulence of different MAC isolates obtained from patients [19, 20]. Our results show that the RgT MAC proliferates in vivo at a much faster rate than the SmT MAC, resulting in higher mortality for infected animals. This could be caused by the higher resistance of the RgT MAC to the antimicrobial arsenal of macrophages than that of SmT MAC as determined by assaying macrophage microbicidal activity. As evidenced in macrophage cultures infected with the same number of SmT and RgT bacilli, the number of RgT MAC recovered from cultured macrophages 2 days postinfection was four times higher than that of SmT MAC (Table 1). There are controversial reports regarding the in vitro and in vivo correlation of virulence. Cultured human monocytes have been used by some to analyse the virulence of mycobacteria [21, 22] but others believe that it might give an
210
erroneous estimate of mycobacterial virulence [15]. Our study shows that varied results could be obtained by using different methods such as the discrepancy in results obtained by the two methods used to assess MAC growth in macrophages (cfu counting and radiolabelled glycerol uptake). We believe that the structural changes in bacilli could lead to differences in metabolic activity, ultimately affecting the outcome of the study. Therefore, our data emphasize the reliability of cfu counts over other methods, particularly when morphological and structural changes in bacilli are involved. Various authors in the past have reported differences in the virulence of various colony morphotypes, but these studies provided epidemiological information rather than models for studying the virulence factors associated with these bacteria. Other studies have used standard strains known to vary in virulence for studying interactions with the cultured macrophages [23, 24]. The major obstacle in conducting a more systematic study was the high frequency of interconversion of these colony forms. We have observed that the SmT MAC could be converted to the RgT MAC progressively with the time of infection in beige mice (Table 3). However, once the RgT variant was obtained, it seldom reverted back to SmT. In addition, the colony morphology and hence the virulence of RgT bacilli could be maintained indefinitely by continuous passage in macrophages. Thus, we have obtained two distinct colony morphotypes of the MAC that originated from the same culture but possess major differences in their morphology, metabolic activity and virulence. Our data indicated significant changes in either the permeability or certain metabolic pathways of the MAC during conversion from the SmT MAC to the RgT MAC, leading to reduced incorporation of compounds such as glycerol, MTT, alamar blue and neutral red. The failure of the RgT MAC to reduce MTT, alamar blue and neutral red might also indicate the possibility of a deficiency in some enzyme systems necessary for the incorporation and metabolization of these compounds. Further studies to delineate the differences in structural or biochemical components between these variants are in progress.
Materials and methods MAC isolate The MAC 101 strain, obtained from Childrens’ Hospital, Los Angeles, CA, U.S.A., was main-
R. G. Kansal et al.
tained on slants of Lowenstein-Jensen medium (Remel, Lenexa, KS, U.S.A.). Culture suspensions were prepared by growing an initial inoculum on Middlebrook 7H11 agar (Difco Laboratories, Detroit, MI, U.S.A.) supplemented with OADC (oleate, albumin, dextrose and catalase) enrichment (Remel) and glycerol (Boehringer Mannheim Biochemicals, Indianapolis, IN, U.S.A.) as described previously [16]. Both types of colonies from these plates were subcultured in Middlebrook 7H9 broth (MBB) (Difco Laboratories) supplemented with OADC enrichment and glycerol and allowed to grow at 37°C on a rotary shaker for 3 days. The culture suspensions were shaken and sonicated for 90 s using a bath-type sonicator (Laboratory Supplies Co., Inc., Hicksville, NY, U.S.A.) and adjusted to the desired concentrations using a Klett Summerson colorimeter (Klett Manufacturing, Brooklyn, NY, U.S.A.) or a spectronic-20 spectrophotometer (Milton Roy, Rochester, NY, U.S.A.). The size of the inoculum was determined by titration, plating and counting CFU on 7H11 agar plates after 2 weeks of incubation at 37°C. The cultures were diluted appropriately to contain equal numbers of bacilli in both SmT and RgT MAC suspensions and kept at −20°C until use.
Animals Female C57 BL/6 J/bgj bgj mice aged 4–6 weeks (weight, 14–16 g) were obtained from Jackson Laboratories (Bar Harbor, ME, U.S.A.). These animals received water and food ad libitum. Animals were used according to a protocol approved by the Institutional Animal Care and Use Committee at the University of Texas M. D. Anderson Cancer Center.
Determination of mycobacterial growth The growth rates of the SmT MAC and the RgT MAC were determined by counting cfu at different times during the MAC culture in broth. Briefly, 10 ml of MBB were inoculated with a known number of bacilli from each of the colony forms and allowed to grow at 37°C on a rotary shaker (160 rpm). Aliquots were taken out at days 0, 1, 2, 4, 6 and 8, and the cfu were determined by plating appropriate dilutions on 7H11 agar plates. The colonies were counted after 2 weeks of incubation at 37°C.
Colony morphology of MAC affects virulence and biochemical properties
[3H]-glycerol uptake assay Both the SmT and RgT MAC were grown in MBB without glycerol but containing [3H]-glycerol (ICN Biochemicals, Irvine, CA, U.S.A.) at a concentration of 50 lCi/ml under shaking conditions at 37°C. One hundred ll aliquots of both MAC types were taken out in microtitre wells on days 0, 1, 2, 4, 6 and 8 of the growth. Fifty microlitres of 5.25% sodium hypochlorite were added to each well, and bacteria were harvested onto fibreglass filters with a semiautomatic cell harvester (Cambridge Technology, Watertown, MA, U.S.A.). The filter-paper discs were dried overnight in separate vials and analysed for radioactivity incorporated into the bacteria by using Scintiverse II (Fisher Scientific, Fair Lawn, NJ, U.S.A.) and a scintillation counter as described previously [17].
MTT reduction assay The MTT reduction assay was done according to a colorimetric method developed previously in our laboratory [25]. Briefly, 10 ml of MBB were inoculated with a known equal number of bacilli from both the SmT or RgT MAC cultures and allowed to grow at 37°C on a rotary shaker (160 rpm). One hundred microlitre aliquots of both types of bacilli were taken out in microtiter plates on days 0, 1, 2, 4, 6 and 8 of growth. Ten microlitres of the 5 mg/ml stock solution of MTT were added to these aliquots and incubated for 3 h at 37°C. At the end of the incubation, 100 ll of the extraction buffer was added to all the wells. After overnight incubation at 37°C, absorbance was measured at 570 nm with a microplate reader (Molecular Devices, Menlo Park, CA, U.S.A.).
Alamar blue assay Alamar blue (Alamar, Sacramento, CA, U.S.A.) is an indicator dye, formulated to measure the proliferation of a variety of eukaryotic cells as well as bacteria quantitatively. A 10% solution of alamar blue in MBB was prepared. A known equal number of both SmT- or RgT-morphotype bacilli were inoculated separately in tubes containing 10 ml of the above solution and incubated at 37°C on a rotary shaker (160 rpm). 100 ll aliquots (in triplicate) were taken out in wells of the microtitre plates on days 0, 1, 2, 4,
211
6 and 8, and the absorbance was read at 570 and 595 nm. The difference in absorbance between 570 and 595 nm was taken as the overall absorbance value.
Neutral red uptake 10 ml of MBB were inoculated with a known, equal number of bacilli from both the SmT or RgT MAC cultures and incubated at 37°C on a rotary shaker (160 rpm). Four hundred microlitre aliquots of each culture were taken out in microfuge tubes on days 0, 1, 2, 4, 6 and 8 of growth and mixed with 8 ll of 0.1% neutral red solution. The mixture was vortexed and incubated for 1 h at 37°C. The cells were then centrifuged at 15 000 g for 15 min and washed twice with phosphate-buffered saline (PBS). 300 ll of the developer (ethanol:acetic acid: water; in v/v ratio 368:3:129) were added to each tube; the contents were mixed, transferred to a 96-well plate, and read at 540 nm.
Restriction fragment length polymorphism analysis To ascertain that the RgT isolate was a variant of the parent strain and not a contaminant, a two-step assay combining a gene amplification step and a restriction fragment length polymorphism analysis, developed previously by Plikaytis et al. [26] was used.
Sample preparation The genomic DNA to be used for amplification was prepared from RgT, SmT as well as reference strains of M. avium (ATCC 35713) and M. intracellulare (ATCC 35761) as described previously [27]. Briefly, the bacterial cell suspensions were prepared in Tris EDTA (TE) buffer and were heat inactivated for 10 min at 80°C, followed by centrifugation at 13 000 g at 4°C for 15 min. The supernatants were then used as source of template DNA for PCR.
Gene amplification Primers corresponding to portions of the M. tuberculosis hsp 65 gene were obtained from
212
GIBCO BRL custom primers of Life Technologies, Gaithersburg, MD, U.S.A. and used to obtain an amplicon on a thermal cycler as described earlier [27].
Restriction enzyme digestion Fifteen microlitres of the amplified reaction mixture was transferred to a fresh tube and digested with either Bst NI (New England Biolabs, Inc., Beverly, MA, U.S.A.) or Xhol (Bethesda Research Laboratories, Gaithersburg, MD, U.S.A.). The products obtained were analysed on 1.5% agarose gels and the DNA bands were visualized by ethidium bromide fluorescence. Gels were photographed with Polaroid type 55 positive/ negative film (Polaroid Corp., Cambridge, MA, U.S.A.).
Growth of the SmT and RgT MAC in cultured macrophages: macrophage harvest and culture Mice were killed by cervical dislocation, and peritoneal macrophages were harvested in cold RPMI 1640 (GIBCO, Gaithersburg, MD, U.S.A.). The cell suspension was centrifuged at 4°C, 200 g for 10 min and resuspended in prewarmed RPMI containing 10% fetal bovine serum, 1% penicillin-streptomycin solution and 1% HEPES. The cell number was determined with a coulter counter (Coulter Electronics, Hialeah, FL, U.S.A.). Cells were then seeded into 96-well plates (Becton Dickinson Co., Lincoln Park, NJ, U.S.A.) at a concentration of 1.5×105 cells/well. After incubation for 2 h at 37°C in 5% CO2, nonadherent cells were removed by washing twice with RPMI 1640 as described elsewhere [17]. Monolayers were then incubated overnight at 37°C in a humidified 5% CO2 incubator.
R. G. Kansal et al.
bacilli and reconstituted with 200 ll of fresh medium. To determine the number of bacilli taken up by the macrophages after 4 h of phagocytosis, half of the wells were lysed at this time and processed for the determination of cfu. The rest of the infected macrophages were incubated further to let the bacilli grow and multiply. After 2 days of incubation, these macrophages were lysed and processed as described below.
Assay of intracellular MAC growth The growth of MAC in macrophages was studied by measuring [3H]-glycerol uptake by viable bacteria or by determining the number of cfu in the macrophage lysate as described previously [17]. Briefly, the infected cells were lysed at the end of the incubation period by removing the medium and replacing it with 100 ll MBB (without added glycerol) followed by the addition of 10 ll 1 N NaOH, 10 ll 1 N HCl (to neutralize NaOH), and 10 ll 5% Tween-80 (Difco Laboratories). For determination of CFU, appropriate dilutions of the macrophage lysate were prepared in PBS, and 100-ll aliquots were plated onto 7H11 agar. Plates were incubated for 12–14 days until the colonies were apparent and cfu were counted. To measure the incorporation of radioactivity by viable bacilli, 1 lCi/well of [3H]-glycerol was added to all the wells and the plates were incubated for 4 days at 37°C. Next, 50 ll of 5.25% sodium hypochlorite were added to all of the wells and bacteria were harvested onto fibreglass filters using a semiautomatic cell harvester. The filter-paper discs were dried overnight in separate vials and analysed for radioactivity incorporated into the bacteria using scintiverse II and a scintillation counter (Packard Instrument Company, Laguna Hills, CA, U.S.A.).
Infection of macrophage cultures
Evaluation of virulence in vivo
After overnight culture, monolayers were washed twice with Excell 300 (JRH Biosciences, Lenexa, KS, U.S.A.) containing 1% HEPES. Monolayers were then infected with 20 ll of culture suspensions containing 1×107 cfus of the SmT and RgT MAC (MAC-to-macrophage ratio, 10:1) for 4 h at 37°C in 5% CO2 [17]. After incubation, infected cells were washed three times with Excell 300 to remove extracellular
The virulence of the SmT and RgT MAC was studied in the beige mice by comparing the mortality of the mice and cfu recovered from the lungs, liver and spleen in the respective groups at various time intervals post-infection as described previously [16]. Groups of 20 mice each received intravenous injections of SmT or RgT MAC cultures containing approximately 5×107 bacilli through the tail vein. These mice
Colony morphology of MAC affects virulence and biochemical properties
213
were observed for mortality up to 60 days postinfection, and the results obtained from mice infected with the RgT MAC were compared with those obtained from mice infected with SmT MAC.
Coordinating Board to RTM and by the National Institute of Health-National Cancer Institute (core) support grant CA-16672 to the Department of Veterinary Medicine and Surgery, The University of Texas M. D. Anderson Cancer Center. We thank Don Norwood for editorial assistance.
Assessment of infection in various organs
References
Three to four animals from both groups were killed on days 1, 8, 15 and 22 post-infection, and vital organs such as the lungs, liver and spleen were removed aseptically. The tissues were weighed and homogenized in a known volume of PBS using an electric homogenizer (Omni International, Inc., Gainesville, VA, U.S.A.). The cfu were determined after plating the appropriate dilutions on 7H11 agar. The plates were incubated at 37°C, and cfu were counted after 2 weeks of incubation. The results were expressed as cfu/g of tissue. To examine the conversion of the SmT morphotype to RgT morphotype, both colony types from each organ were counted and the conversion percentage was calculated.
Bacteraemia and blood profile Blood was collected post-infection from the retro-orbital plexus of the infected animals at the indicated time periods, and a 100-ll aliquot was plated on the MBB agar plates to find out the extent of bacteraemia in animals infected with SmT and RgT bacilli. Other blood parameters such has haemoglobin, haematocrit and the number of red blood cells were also examined on day 35 post-infection when the infection was well established.
Statistical analysis The results were expressed as the mean ±SD of triplicates from one representative experiment; all experiments were performed three times with similar results. The level of significance was assessed by Student’s t-test.
Acknowledgements This work was supported by grants ATP-D 000015084 and ATP 000015091 from the Texas Higher Education
1 Horsburgh CR Jr., Selik RM. The epidemiology of disseminated nontuberculous mycobacterial infection in the acquired immunodeficiency syndrome (AIDS). Am Rev Resp Dis 1989; 139: 4–7. 2 Macher AM, Kovacs JA, Gill V et al. Bacteremia due to Mycobacterium avium-intracellulare in the acquired immunodificiency syndrome. Ann Intern Med 1983; 99: 782–5. 3 Murray JF, Felton CP, Garay SM et al. Pulmonary complications of the acquired immunodeficiency syndrome. N Engl J Med 1984; 310: 1682–8. 4 Young LS. Mycobacterium avium complex infection. J Infect Dis 1988; 157: 863–7. 5 Pattyn SR, Hermans-Boveroulle MT. Dissociation in M. avium. Pneumonology 1970; 142: 119–25. 6 Inderlied CB, Kemper CA, Bermudez LE. The Mycobacterium avium complex. Clin Microbiol Rev 1993; 6: 266–310. 7 Benson CA. Treatment of disseminated disease due to M. avium complex in patients with AIDS. Clin Infect Dis 1994; 18(suppl 3): S237–42. 8 Pattyn SR. A study of group 111 non chromogenic mycobacteria. Correlation of chicken virulence with other in vitro characters among 20 strains. Zeitsch fu¨r Tuberk 1967; 127: 41–6. 9 Moehring JM, Solotorovsky MR. Relationship of colonial morphology to virulence for chickens of Mycobacterium avium and the non-photochromogens. American Review of Respiratory Disease 1965; 92: 704–13. 10 Schaefer WB, Davis CL, Cohn ML. Pathogenicity of transparent, opaque and rough variants of Mycobacterium avium in chickens and mice. Am Rev Resp Dis 1970; 102: 499–506. 11 Anz W, Meissner G. Comparative virulence test in chickens with transparent and opaque strains of avian mycobacteria of different serotypes. Zentralblatt Fur Bakteriologie, Parasitenkunde, Infektionskranheiten Und Hygiene 1972; 221: 334–42. 12 Reddy VM, Luna-Herrera J, Gangadharam PRJ. Pathobiological significance of colony morphology in Mycobacterium avium complex. Microb Pathog 1996; 21: 97–109. 13 Gangadharam PRJ. Beige mouse model for Mycobacterium avium complex disease. Antimicrob Agents Chemother 1995; 39: 1647–54. 14 Kuze F, Uchihira F. Various colony formers of Mycobacterium avium-intracellulare. Eur J Resp Dis 1984; 65: 402–10. 15 Pedrosa J, Florido M, Kunze ZM et al. Characterization of the virulence of Mycobacterium avium Complex (MAC) isolates in mice. Clin Exp Immunol 1994; 98: 210–16. 16 Mehta RT. Liposome encapsulation of clofazimine reduces toxicity in vitro and in vivo and improves therapeutic efficacy in the beige mouse model of
214
17
18
19
20
21
disseminated Mycobacterium avium-M. intracellulare complex infection. Antimicrob Agents Chemother 1996; 40: 1893–902. Gomez-Flores R, Tucker SD, Kansal RG, Tamez-Guerra R, Mehta RT. Enhancement of antibacterial activity of clofazimine against Mycobacterium avium-intracellulare complex infection induced by IFN-c is mediated by TNF-a. J Antimicrob Chemother 1997; 39: 189–97. Kansal RG, Gomez-Flores R, Sinha I, Mehta RT. Therapeutic efficacy of liposomal clofazimine against Mycobacterium avium complex in mice depends on size of initial inoculum and duration of infection. Antimicrob Agents Chemother 1997; 41: 17–23. Gangadharam PRJ, Perumal VK, Jairam BT, Podapati R, Taylor RB, LeBrecque JF. Virulence of Mycobacterium avium complex strains from acquired immune deficiency syndrome patients: relationship with characteristics of the parasite and host. Microb Pathog 1989; 7: 263–78. Reddy VM, Parikh K, Luna-Herrera J, Falkinham JO, Brown S, Gangadharam PRJ. Comparison of virulence of Mycobacterium avium complex (MAC) strains isolated from AIDS and non-AIDS patients. Microb Pathog 1994; 16: 121–30. Toba H, Crawford JT, Ellner JJ. Pathogenicity of Mycobacterium avium for human monocytes: absence of macrophage activating factor activity of gamma interferon. Infect Immun 1989; 57: 239–44.
R. G. Kansal et al. 22 Crowle AJ, Tsang AY, Vatter AE, May MH. Comparison of 15 laboratory and patient-derived strains of Mycobacterium avium for ability to infect and multiply in cultured human macrophages. J Clin Microbiol 1986; 24: 812–21. 23 Sarmento AM, Appelberg R. Relationship between virulence of Mycobacterium avium strains and induction of tumor necrosis factor alpha production in infected mice and in in vitro-cultured mouse macrophages. Infect Immun 1995; 63: 3759–64. 24 Sarmento AM, Appelberg R. Involvement of reactive oxygen intermediates in tumor necrosis factor alphadependent bacteriostasis of Mycobacterium avium. Infect Immun 1996; 64: 3224–30. 25 Gomez-Flores R, Gupta S, Tamez-Guerra R, Mehta RT. Determination of MICs for Mycobacterium avium-M. intracellulare complex in liquid medium by a colorimetric method. J Clin Microbiol 1995; 33: 1842–6. 26 Plikaytis BB, Plikaytis BD, Yakrus MA et al. Differentiation of slowly growing Mycobacteria species, including Mycobacterium tuberculosis, be gene amplification and restriction fragment length polymorphism analysis. J Clin Microbiol 1992; 30: 1815–22. 27 Telenti A, Marchesi F, Balz M, Bally F, Bottger EC, Bodmer T. Rapid identification of Mycobacteria to the species level by polymerase chain reaction and restriction enzyme analysis. J Clin Microbiol 1993; 31: 175–8.