SEARCH

SEARCH BY CITATION

Keywords:

  • cytometry;
  • fluorescence;
  • malaria diagnosis;
  • tuberculosis diagnosis;
  • tuberculosis antimicrobial susceptibility testing

Abstract

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED

In many resource-poor areas, the HIV/AIDS epidemic coexists with epidemics of two much older diseases, malaria and tuberculosis, and the three diseases together kill approximately six million people per year. Although HIV/AIDS is treatable, but not curable, many if not most cases of malaria and tuberculosis (TB) can be cured if diagnosed correctly and promptly. The diagnosis of both malaria and TB is cell-based, typically made by microscopy of stained smears of blood (malaria) and sputum (TB). Cytometry has been shown to be effective for diagnosis of both conditions; however, conventional cytometers have been too complex and costly to be widely applied. It is likely that a newly developed small, simple, robust, inexpensive, energy-efficient low-resolution fluorescence image cytometer, employing a light-emitting diode for excitation and a megapixel digital camera chip for detection, could be used in resource-poor areas for malaria and TB diagnosis and for rapid (24–48 h) determination of antimicrobial susceptibility of Mycobacterium tuberculosis. © 2008 Clinical Cytometry Society

AIDS was the first and remains the most significant disorder to be defined by results of fluorescence flow cytometry. When the original report by Gottlieb et al. (1) was published in 1981, instruments with the capability of measuring CD4 antigen expression on T lymphocytes by immunofluorescence had been available for less than a decade, and the necessary monoclonal antibodies had only recently reached the market. The absolute count of CD4+ T cells in peripheral blood was quickly identified as a valuable prognostic measurement in HIV-infected patients (2), and HIV-related CD4 counting provided, literally and figuratively, a “killer application” for fluorescence flow cytometers (3), at least in affluent countries.

We now know that AIDS is a new disease, caused by viruses that made the leap from other primates to man somewhere in Africa around the middle of the 20th Century (4), and it is in Africa, and in other resource-poor areas of the world, that the disease has had the greatest impact on public health. In many such areas, the HIV/AIDS epidemic coexists with epidemics of two much older diseases, malaria (5) and tuberculosis (TB) (6), and the three diseases, which exacerbate one another (7–10), together kill approximately six million people per year.

Although governmental and nongovernmental agencies are now working together to provide antiretroviral therapy to HIV-infected individuals in resource-poor countries, the available drugs, even with negotiated discounts, are relatively expensive, and do not cure the infection. By contrast, many if not most cases of malaria and TB can be cured using relatively inexpensive drugs, if the disease is diagnosed correctly and promptly. In resource-poor settings, diagnosis of both malaria and TB is cell-based, and depends on detection of the causative agent using transmitted light microscopy to examine a stained smear of blood (in the case of malaria) or sputum (in the case of TB). Although cytometry has been applied to diagnosis of both malaria and TB and to rapid determination of drug susceptibility and resistance in TB, the cost and complexity of both flow and imaging cytometry have until recently precluded serious consideration of cytometry for routine application, even in developed countries. In this paper, we will summarize what has been done to date and attempt to make a case for the use of simple, inexpensive low-resolution fluorescence imaging cytometry (11, 12) for affordable diagnosis of malaria and TB and for detection of drug resistance in TB.

MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED

Malaria

Malaria kills as many as three million people annually, most of them children, and hundreds of millions of episodes of clinical illness occur each year. Most mortality is attributable to the severest form of the disease, caused by the protozoan parasite Plasmodium falciparum; P. vivax, P. malariae, and P. ovale are substantially less likely to cause life-threatening illness.

Descriptions of malaria appear in ancient Sumerian and Chinese writings; scientific detective work places the origin of the disease in prehistoric Africa, and native populations in South America, where the disease was introduced by European conquistadores in the 1500s, found the first effective therapy in the form of cinchona bark, from which the active ingredient quinine was later isolated. It was only in 1880 that Alphonse Laveran detected the parasite in red blood cells (RBCs), without benefit of either an oil immersion microscope objective (introduced in 1884) or staining (5, 13, 14); his work won him the Nobel Prize for Medicine in 1907.

The genome of P. falciparum was sequenced in 2002 (15); the haploid genome is ∼23 Mbp in size and contains 80% adenine + thymine (A+T), the highest such percentage found in any organism analyzed to date. P. vivax, P. malariae, and P. ovale appear to have haploid genome sizes of about 30 Mbp and contain about 60% A+T (16, 17). All forms of Plasmodia except the zygote, which forms after fertilization in the mosquito midgut, are haploid; that is, the nuclei have a “1C” DNA content, although nuclear replication occurs during the development of later stages of the parasite in human RBCs and liver cells. Schizonts, the latest stage, may have DNA content as high as 24C. Gametocytes, found in peripheral blood, have DNA content higher than 1C, but lower than 2C (18).

Tuberculosis

Evidence of TB in humans has been found in Egyptian mummies and in European Stone Age burials dating back more than 6,000 years. Robert Koch reported the discovery of the causative organism of TB, Mycobacterium tuberculosis (MTB), in 1882, and was awarded the Nobel Prize for his work in 1905. Koch noted that MTB was slow-growing and difficult to stain (6, 13). The MTB genome sequence was reported in 1998 (19); the organism contains ∼4.4 Mbp of DNA, with (A+T) about 33%.

The World Health Organization (WHO) estimates (20) that 8.8 million cases of TB occurred in 2005, with 3.9 million of these representing the most infectious, smear-positive, form. Of the estimated 1.6 million deaths, 195,000 occurred in HIV-infected individuals. Although the global incidence rate of TB peaked around 2002, and appears now to have stabilized or begun to decline, the current slow decline in incidence rates per capita is offset by population growth, and the number of new cases per year is still increasing.

Also according to WHO (21), over 424,000 of the 8.9 million TB cases that occurred in 2004, the last year for which data on resistance are available, were multidrug resistant (MDR), i.e., resistant to isoniazid (INH) and rifampin; 62% of these occurred in China, India, and the Russian Federation. Among MDR cases, ∼10% are extensively drug resistant (XDR) (22, 23), i.e., MDR plus resistant to any fluoroquinolone and an injectable drug (amikacin, capreomycin, or kanamycin).

Although the American media and public were recently made aware of the dangers of XDR-TB through accounts of the odyssey of one individual misdiagnosed as infected with the disease (24), XDR-TB poses the greatest threats to public health in the resource-poor areas mentioned above, and the WHO has recently emphasized the need for inexpensive, rapid, and simple TB diagnostic and antimicrobial susceptibility tests suitable for use in such areas.

Malaria, TB, and the History of Cytometry

From the vantage point of the 21st century, it is easy to forget that, until the end of World War II, even advanced microscopy laboratories had little more in the way of resources than are available in “resource-poor” countries today. The compound microscope with which Robert Hooke visualized “cells,” actually the remnants of cell walls in thinly sliced cork, in 1663 (25) had a magnifying power of less than 50×, and Antoni van Leeuwenhoek's discoveries of blood and sperm cells and bacteria in the late 1600s and early 1700s were made using simple lenses, which he himself made and which, remarkably, achieved magnifications almost 10 times higher. Until photography was developed, in the mid-1800s, there were no objective means of recording the results of observations made using microscopes. Also, although van Leeuwenhoek had applied saffron to muscle tissue in an attempt to improve visibility of subcellular structures, it was not until the 1870s that staining of cells and tissues came into wide use, with Paul Ehrlich playing a pivotal role (26, 27).

By 1850, the cell theory had become accepted, and, with the aid of improved compound microscopes developed by Zeiss and other manufacturers, Rudolf Virchow was attempting to establish the cellular basis of anatomic pathology. Although Virchow was able to detect and define leukemia based on observations of unstained blood, it was difficult to distinguish among different types of leukocytes [white blood cells (WBCs)] in unstained specimens. While studying medicine during the 1870s, Ehrlich, whose cousin, Karl Weigert, had been Virchow's assistant, experimented with a variety of newly synthesized aniline dyes, in search of dyes and combinations that would facilitate the discrimination of different cell types. He demonstrated that combinations of acidic and basic dyes, which he called neutral stains, could produce differential staining of WBCs and that at least some types of bacteria could readily be stained by the basic dye methylene blue.

The synthetic dyes themselves were the serendipitous result of a failed attempt to synthesize quinine, a proven treatment for malaria but also a relatively scarce and expensive natural product. In 1856, William Perkin, studying at the Royal College of Chemistry in London under August von Hofmann, subjected aniline to distillation and oxidation, with the end result not the expected quinine, which is colorless, but the purple dye later known as mauve. Although von Hofmann was initially unimpressed, Perkin soon found that the burgeoning British textile industry would buy large quantities of the inexpensive synthetic dye, which was more intense and color fast than almost all of the natural dyes that had been used to date. Both Queen Victoria and Napoleon III's Empress Eugénie favored mauve, which, thanks to the resulting popular demand and the efforts of other chemists in Great Britain, France, and Germany, was soon joined in the marketplace by many other synthetic dyes.

After considerable experimentation, Koch had succeeded in staining MTB with methylene blue, but only after smears on slides had been exposed to alkaline solutions of the dye for many hours. In 1882, Ehrlich joined Koch, and soon found that MTB could be stained more rapidly if aniline water was substituted for Koch's potassium hydroxide, and that, once stained, the organism was “acid-fast,” that is, it retained dye after the stained smear was washed with strong acid solutions. By 1883, Ziehl had replaced aniline water with phenol, and Neelsen had demonstrated that carbol fuchsin (basic fuchsin in a phenolic solution) provided specific acid-fast staining. The Ziehl-Neelsen stain, with methylene blue used as a counterstain after an acid-alcohol wash, has remained the standard for detection of MTB using transmitted light microscopy for over 120 years. MTB and other acid-fast organisms appear reddish-pink; other types of bacteria, cell nuclei, and other debris are stained blue.

The acid-fast character of mycobacteria is due to the presence of mycolic acids (28–31), responsible for the “waxy” consistency of the cell wall of these organisms. When originally isolated and described by Stodola et al. in 1938 (28), “mycolic acid,” which was noted to stain with carbol fuchsin and retain the dye after acid-alcohol washing, was thought to be a single, ether-soluble, high molecular weight hydroxy acid; it was later established (29) that the material contained a variety of β-hydroxy fatty acids with long α-alkyl side chains, and it is now known (30–32) that mycolic acids, in the plural, are a family of several hundred intimately related discrete chemical structures, differing in composition and proportion from species to species and, in some cases, from strain to strain.

In 1938, the same year that mycolic acid was first described (28), Hagemann (33) reported that acid fast fluorescent staining of MTB could be achieved using the blue- or blue-green- (∼430–490 nm) excited, green- (∼530 nm) fluorescent dye auramine O. This dye, used either alone or in combination with rhodamine B, remains the mainstay for clinical diagnosis of MTB by fluorescence microscopy of sputum smears. A number of studies, and a recent extensive analysis by Steingart et al. (34) in which they are cited, support the conclusion that fluorescence microscopy with auramine O or auramine-rhodamine is more sensitive than transmitted light microscopy with the Ziehl-Neelsen stain. The fluorescence method has the advantage that it allows an observer to scan a slide under lower magnification (250–450×, without oil immersion) than is necessary with transmitted light (1,000×, oil immersion). Two obvious disadvantages of fluorescence microscopy are the relatively high cost of apparatus, which may soon be mitigated by the adoption of light-emitting diodes (LEDs) for fluorescence excitation (35–37), and the frequent occurrence of nonspecific fluorescent staining of cellular debris and bacteria other than Mycobacteria, making it necessary that observers be well-trained.

Although the dye mixtures used in Ehrlich's original neutral stains for leukocytes (WBCs) in blood smears were sufficient for the discrimination of mononuclear cells and different types of granulocytes, they did not stain malaria parasites, discovered in 1880 in unstained blood, inside or outside of RBCs. By 1891, Chenzinsky, Malachowski, and Romanowsky had independently found that aged or alkali-treated methylene blue, combined in a neutral stain with the acid dye eosin, would produce reddish-purple staining of WBC nuclei and of malaria parasites in blood smears. The stain combination was later modified in various ways by Nocht, Jenner, Reuter, Leishman, Wright, and Giemsa; Wright's stain became widely used for differential leukocyte counting in the United States, and Giemsa's, first described in 1904 (26, 38), and also widely used for hematology, remains the standard stain for diagnosis of malaria in blood films by transmitted light microscopy.

Fluorescence microscopy using acridine orange (39–41) and other nucleic acid stains (42–44) has been reported to compare in accuracy with transmitted light microscopy for malaria diagnosis (45), and may require less time and a less-skilled observer, but, as is the case with TB diagnosis, cost of the instrument remains an issue.

MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED

Detecting and Counting the Pathogens

Even when fluorescence microscopy is used, both TB and malaria diagnosis by human observers present the problem of examining a large enough volume of specimen in a reasonable time to insure detection and precise quantification of small numbers of relatively rare cells of interest. It is this problem that has historically been solved by cytometry; if malaria and TB were of more concern in affluent countries, cytometric techniques would likely have supplanted microscopy of stained smears for diagnosis, just as they have done for differential WBC counting and reticulocyte counting in clinical hematology.

As “Student” showed in 1907, cell counting is governed by what are now called Poisson statistics (46). If n objects of interest are actually counted, the standard deviation of the count can be no less than √n, and the relevant measure of precision, the coefficient of variation (CV), expressed as a percentage, can be no less than 100(√n/n). As will be shown in the following sections, diagnosis of both malaria and TB from smears by microscopy under standard conditions typically yields highly imprecise estimates of the number of organisms present and may fail to detect small numbers of pathogens on slides. Cytometry is obviously advantageous in that it allows analysis of a larger volume of a specimen, thereby improving both accuracy and precision of diagnosis, but this is only possible if the instrument can discriminate between pathogens and other objects in the specimen at least as accurately as can a human observer.

Malaria Diagnosis from Blood Smears and Cytometric Alternatives

The diagnosis of malaria depends on the detection and counting of parasites on Giemsa-stained thick and/or thin blood smears. It is reported (47) that an experienced microscopist can detect levels of 50 parasites/μL, or (assuming the RBC count to be 5,000,000/μL) 0.001% RBC infected, whereas an observer with less training may not be able to detect fewer than 500 parasites/μL, or 0.01% RBC infected.

Although some parasitized RBCs are sequestered from the circulation, notably in P. falciparum malaria, parasite density, i.e., the number of parasites per unit volume of blood, is generally accepted as the best measure of the severity of infection. Determination of parasite density requires some measure or estimate of the volume of blood examined; however, measurements of density based on analyses of Giemsa-stained smears are almost always imprecise, often inaccurate, and not at all well standardized (48–50). Since it is difficult for an observer to know the actual volume of blood he or she or he has examined in a smear, it is common practice to look at enough fields on the slide to count 200 or more WBCs. If the WBC count is known, the parasite density can then be calculated, but, in the absence of a WBC count, it is simply assumed that the count is some arbitrary value in the normal range, e.g., 8,000/μL, thereby decreasing the accuracy of the reported density.

Poisson statistics dictate that keeping the CV of the density measurement under 10% requires that at least 100 parasites be counted. If this level of precision is needed at the limit of sensitivity, 50 parasites/μL, it becomes necessary to analyze at least 2 μL of blood. Although a typical thick smear contains between 10 and 25 μL of blood, the volume of blood normally examined by microscopy is substantially less. If the WBC is 8,000/μL, and one counts an area of the slide in which a total of 500 WBCs are found, the volume of blood examined is only 0.0625 μL, and, at a parasite density of 100/μL, only six parasitized cells would be counted on average. Even if an accurate WBC count were used in the calculation, the resultant CV could therefore be no lower than 40.8%. This assumes insignificant loss of parasites as a result of the distilled water wash normally applied to the thick smear to lyse RBCs; it has, however, been suggested (48) that as many of 80% of parasites may be lost during this step.

Although the Giemsa stain dates from 1904, the mechanism by which two dyes combine to yield three colors was not understood until the 1980s (51–53). The anionic dye eosin Y readily binds to basic groups on proteins in cells and tissues, staining them pink; cationic azure dyes, formed by the oxidation of methylene blue and like it, incorporating a tricyclic thionin ring, bind to nucleic acids. Cytoplasmic RNA is typically stained blue; however, nuclei, in which DNA is concentrated, appear purple. The Giemsa effect can be achieved using two pure dyes, eosin Y and azure B; the purple color results from the formation of complexes of eosin Y and DNA-bound azure B (51–53). The stain colors Plasmodium nuclei reddish purple, and cytoplasm bluer; it is not, however, readily possible to use measurements of the absorption of the component dyes to derive quantitative measures of DNA or RNA content.

Only a few dozen of the over 85,000 publications in the cytometry literature deal with the diagnosis, treatment, or biology of malaria (54). The efficacy of flow cytometry for detection, characterization, and counting of human and animal malaria parasites using fluorescent nucleic acid dyes was first noted in the 1970s, and publications continue to emerge (54–58). Unfortunately, although it was shown in the mid-1980s that different developmental stages of P. falciparum could be identified by flow cytometry based solely on measurements of nucleic acid (DNA and RNA) content, without recourse to morphologic information (59, 60), most people in both the malariology and cytometry communities remain unaware of this highly significant finding.

In clinical diagnosis, it is important to distinguish between malaria caused by P. falciparum and malaria due to the other species. In the former, peripheral blood contains a preponderance of haploid and near-haploid forms; in the latter, later stages of the parasites, with DNA content as much as 24 times higher, are also common in peripheral blood. Flow cytometry of blood stained with DNA-selective fluorescent stains, e.g., Hoechst 33258, has already been shown capable of detecting both individual haploid forms and later stages of P. falciparum (later stages are present in cultures) and other species (54, 56). In clinical samples, it should therefore be possible both to detect malaria and to differentiate P. falciparum malaria from other human malarias based on the distribution of DNA content. Although blood does contain small fragments of DNA, they are typically only a few hundred base pairs in length, much smaller than a Plasmodium genome, or even a bacterial genome (fewer than 10 Mbp), and thus would not interfere significantly with cytometric detection of parasites. WBC nuclei contain orders of magnitude more DNA and also would not interfere with parasite detection; moreover, cytometry could readily provide both a WBC count and a parasite count from the same volume of sample. Thus far, however, the sensitivity of flow cytometry has not typically reached the 50 parasites/μL level, probably because it has been difficult, using stains that bind to both DNA and RNA, to resolve haploid or near-haploid parasites from reticulocytes and background debris.

Another approach to flow cytometric diagnosis of malaria relies on detection of hemozoin, a breakdown product of hemoglobin produced by the parasite, in WBCs (61–64). The measurement requires determination of the magnitudes of polarized and depolarized side scatter, rather than fluorescence, and can be implemented in the Cell-Dyn series of hematology counters (Abbott Diagnostics, Santa Clara, CA). The number of hemozoin-bearing monocytes and granulocytes is related to the severity of the disease, but is also affected by the immune status of the patient, suggesting that hemozoin measurements may be less useful than direct detection of parasites for primary diagnosis of malaria.

TB Diagnosis from Sputum Specimens: CurrentState of the Art

It is now firmly established that fluorescence microscopy using auramine O with or without rhodamine B is more sensitive than transmitted-light microscopy for TB diagnosis (34), and that concentration of sputum using chemicals (e.g., a sodium hydroxide [NaOH] and N-acetylcysteine mixture or hypochlorite bleach) followed by centrifugation or overnight sedimentation improves sensitivity for both transmitted light and fluorescence microscopy (65). Nonspecific auramine O fluorescent staining of bacteria other than Mycobacteria and of cell debris remains a problem, especially when examining unprocessed sputum, because the observer must use morphologic criteria to discriminate MTB organisms and microcolonies from nonspecifically stained material. If auramine O were, as stated in many texts (e.g., Ref.66), selective for mycolic acid, one would not expect nonspecific staining with the dye to occur, since carbol fuchsin, demonstrated to produce acid-fast staining of mycolic acid (28), does not produce significant nonspecific staining.

In order to develop a relatively simple cytometric procedure for TB diagnosis using auramine O, the specificity of staining must be improved; this requires reexamination of the purported staining mechanism (36). The original experiments reporting acid-fast staining of mycolic acid by auramine O (67) were, like those showing acid-fast staining with carbol fuchsin (28), done with solid purified material on slides, rather than with bacteria; staining was not shown to be selective for mycolic acid in either case. We have confirmed by spectrofluorometry that, as has been known for some time, the fluorescence of auramine O is enhanced in the presence of DNA (68) and RNA (69), whereas we found that the fluorescence of the dye in organic solvents in which mycolic acid is dissolved is not substantially different from fluorescence of the dye in solvent alone. Moreover, confocal microscopy of auramine O-stained MTB reveals that most of the fluorescence comes from within cells rather than from the cell wall (36). This suggests that most auramine O staining of MTB is due to binding of the dye to DNA and RNA. Auramine O is relatively hydrophobic, and we and others (70–72) have noted that M. smegmatis, M. avium, and MTB readily take up other hydrophobic fluorescent dyes that stain nucleic acids, whereas Mycobacteria are relatively impermeant to more water-soluble dyes.

The hypothesis that auramine O binds predominantly to nucleic acid is also supported by the observations of Pina-Vaz et al. (71, 72) that preincubation with the DNA- and RNA-binding dye propidium iodide (PI) prior to auramine O staining of sputum smears greatly decreases nonspecific staining, allowing MTB to be detected reliably using a laser scanning cytometer (CompuCyte Corporation, Cambridge, MA). They reported that PI, which is considerably more hydrophilic than auramine O, did not enter or stain either unfixed or heat-killed MTB. The affinity of PI for DNA is several hundred times that of auramine O; thus, it is plausible that PI binds to and blocks the binding of auramine O to DNA and RNA not contained in Mycobacteria, leaving DNA and RNA within the organisms free to form fluorescent acid-fast complexes with auramine O, which is not removed by an acid-alcohol wash because the mycolic acids in the cell wall are relatively impermeant to strongly polar solvents (73, 74). Other relatively hydrophilic nucleic acid-binding compounds, fluorescent or otherwise, with higher binding affinities than auramine O, e.g., methylene blue and some of its derivatives, should also reduce nonspecific staining and therefore be effective as counterstains. Although rhodamine B is commonly used in combination with auramine O for fluorescent staining of MTB, there is no obvious basis in dye chemistry for this practice. We have examined positive and negative sputum slides prepared by Dr. Pina-Vaz using the PI/auramine O technique, and by Dr. Stewart Lipton using a rapid stain (Scientific Device Laboratory, Des Plaines, IL) (75) with a counterstain we believe to be a methylene blue derivative, and have found little or no auramine O-stained material on negative slides.

The cytometry literature contains only a few dozen publications on cytometry of Mycobacteria. Laser scanning cytometry (72) and automated fluorescence microscopy (76, 77), although capable of detecting MTB on auramine O-stained slides, require relatively large, complex, and expensive equipment, and neither these techniques nor flow cytometry, which requires equipment of similar complexity and cost, could be considered practical for use in resource-poor areas. Flow cytometry would, moreover, require extensive preprocessing of sputum and, if an acid-fast staining technique were to be used, it would probably need a wash step, making sample preparation more elaborate.

“Smear-negative” TB, in which MTB is not detected on the sputum smear although sputum culture may be positive, is a particular problem in patients with HIV/AIDS (78). In many cases, positive cultures of MTB are obtained when smears are negative. Both textbooks and journal articles (e.g., Refs.79,80) frequently explain the discrepancy by statements such as “about 5,000–10,000 acid-fast bacilli per mL of sputum must be present for detection by smear, whereas culture requires only 10–100 viable organisms.”

The classic text Toman's Tuberculosis (81), published by the WHO and available free online, provides data as well as a useful theoretical discussion. Toman's points out that a slide containing 10 μL of sputum spread out over a 10 × 20 mm2 area (200 mm2) is comprised of ∼10,000 oil-immersion fields (100× objective, 10× eyepiece, the configuration commonly used for examining Ziehl-Neelsen stained smears). If there are 5,000 organisms or microcolonies/mL sputum, there will be only 50 on the entire slide, or one per 200 fields. Since, in practice, an observer can scan no more than a few hundred fields, it is unlikely that more than a few organisms will be seen, and reasonably likely that none will. Fluorescence microscopy, using a 25× objective, which increases the size of an individual field by a factor of 16, reduces the number of fields per slide to 625, increasing the likelihood that at least a few organisms will be detected in the time typically available for an observer to examine the slide.

A 1973 paper (82) reported good agreement between numbers of MTB detected on slides and colonies obtained by culture of processed sputum specimens at levels above a few thousand organisms/mL. More recent experiments in which MTB or M. smegmatis was spiked into negative sputum samples at known concentrations (83, 84) reported a detection limit of 250–300 organisms/mL when 400 to 500 oil-immersion fields were examined by transmitted light microscopy of Ziehl-Neelsen stained smears. It is likely that lower concentrations could be detected using fluorescence microscopy, but not clear that human observers in most clinical settings could routinely scan an entire slide, even using fluorescence microscopy. Although there exist reagents more specific to MTB than acid-fast stains, e.g., monoclonal antibodies (85, 86), which have been used in flow cytometry (86), and probes for ribosomal RNA (87–89), using more specific reagents cannot improve sensitivity of detection when the concentration of organisms is so low that the volume of specimen examined contains none.

DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED

Most methods of determination of the antimicrobial susceptibility of MTB rely on detection of growth of the organisms in culture in the presence and absence of drugs; inoculation of cultures for susceptibility testing is usually not initiated until a clinical isolate has been obtained in pure culture. Unlike many common bacterial pathogens, which replicate in a fraction of an hour, MTB has a relatively long generation time, usually given as ∼18 h. It is often assumed that a bacterial colony visible to the naked eye contains at least a million organisms; growth to this population size from a single organism requires 20 generation times, which, for MTB, would be 15 days. A recent study comparing methods of diagnosis and susceptibility determination (90) reported that median time to culture positivity using the classical Löwenstein-Jensen solid medium culture method was 26 days and median time to sensitivity results was 68 days, whereas, using modern automated culture methods, which detect growth based on metabolic activity rather than by observation of organisms, median times were 13 days to culture positivity and 22 days to susceptibility results.

Under ideal conditions, effects of drugs on microorganisms can be determined in less than a generation time, using indicators of the physiologic state of the organisms. In 1985, Resnick et al. (91) estimated the membrane potential of M. smegmatis from rhodamine 123 fluorescence measurements made with a flow cytometer, and speculated that the technique could be used for rapid assessment of drug susceptibility. We (92, 93) have investigated effects of antibiotics on both membrane potential and permeability of Staphylococcus aureus and other species, noting that, in at least one case, activity of an antimicrobial was reflected by membrane potential change in less than 15 min. In work with the highly infectious MTB, which requires Biosafety Level 3 containment in developed countries, it is, however, more practical to let a cytometer perform the simpler and safer task of counting cells or colonies that are known to be nonviable before they come in contact with either the cytometer or its operator.

A series of studies (94–101) published since 1995 by Ronald Schell and his colleagues at the University of Wisconsin established that flow cytometry can be used to determine susceptibility of MTB to all first-line and several second-line drugs within 24–36 h; other groups (71, 102–104) have obtained similar results. Reis et al. (103) noted that their work was entirely done in a developing country (Brazil). The most recent work on flow cytometric determination of MTB susceptibility has, for reasons noted above, focused on “safe” methods, in which microorganisms are killed, e.g., by exposure to 1% paraformaldehyde for 40 min (97), before being brought to the cytometry facility.

A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED

In modern cytometry, the use of specific reagents makes it unnecessary to collect morphologic information from cells, eliminating the need for high-resolution imaging. A flow cytometer can make sensitive and precise measurements of light scattered or emitted by individual cells at a dozen or more different wavelengths, even though the cell images obtained are typically not in focus. Based on prior experience with both flow and image (105, 106) cytometry, and more recent theoretical (11) and experimental (12) work, we concluded that it should be possible to build a small, robust, energy-efficient, and extremely inexpensive fluorescence image cytometer that could perform essentially the same measurements now done for TB diagnosis by laser scanning cytometry (72) and automated microscopy (76, 77) and for rapid (24–48 h) antimicrobial susceptibility determination of MTB by flow cytometry (71, 94–104).

The complexity and cost of the apparatus are minimized by using a digital camera to simultaneously capture low-magnification (0.5× to 2×) images of all of the cells in a specimen. Fluorescence excitation is provided by a high-intensity LED, similar to those that have recently been used in fluorescence microscopes (35–37, 107). A similar low-resolution fluorescence imaging system, capable of sensitive fluorescence measurements, was described in 1994 (108); at that time, however, both the digital camera and the laser used for illumination cost thousands of dollars, as would an arc lamp, the only feasible alternative light source. By 2004, when we began building hardware, high-intensity LEDs, and the optics needed to use them efficiently as fluorescence excitation sources for microscopy and cytometry, were available for only a few dollars (107), and suitable digital cameras could be bought for a few hundred dollars (12). Our minimalist image cytometer prototype is shown in Figure 1.

thumbnail image

Figure 1. The minimalist image cytometer prototype (October 2007 version).

Download figure to PowerPoint

The light source for the substage illuminator is a 3W Luxeon III Star “Royal blue” (center wavelength ∼450 nm) LED (LumiLEDs, San Jose, CA; http://www.lumileds.com), which can be powered by a 1,000 mA current regulated supply incorporating 5 D batteries, two resistors, two capacitors, and a voltage regulator IC. A narrow-beam FLP series molded plastic lens (Fraen S. r. l., Cusago, Italy, a division of Fraen Corp., Reading, MA; http://www.fraen.com) (107) is mounted in an integral plastic holder that fits directly over the LED circuit board. The lens incorporates reflective and refractive optics, and captures ∼80% of the total power emitted by the LED. As an excitation filter, we use a 25-mm-diameter, 50-nm bandwidth interference filter with a 445-nm center wavelength (Chroma Technology, Rockingham, VT; http://www.chroma.com), blocked to O.D. 5.0 or higher from 300 to 410 nm and from 485 to 850 nm. A 25-mm-diameter piece of 5° holographic diffuser sheet (Physical Optics Corp., Torrance, CA; http://www.poc. com) is mounted atop the filter. The condenser lens is plano-convex with a 60 or 75 mm focal length; it delivers at least 50 mW of illumination, uniform to 20%, to a 6-mm-diameter area of the specimen.

Two 2.5× microscope objectives (Rolyn Optics, Covina, CA) are used front-to-front for fluorescence collection, forming a 1× image, with the emission filter (527 nm, 30 nm band pass; Chroma) placed between the lenses. Because the cytometer operates at low resolution and has a large field of view (currently ∼6.4 × 5.1 mm2), it needs neither stage motion nor focus controls and thus has no moving parts.

The cameras we use incorporate a 1.3 megapixel (1280 × 1024) 1/2" CMOS chip with 5.2 μm2 pixels and 10-bit monochrome output (Micron Technology, Boise, ID); the resolution of the optics (>100 line pairs/mm) is compatible with the pixel size. Thus far, we have used cameras that interface the chip to a PC via a USB port. Although the price of these cameras has dropped substantially over the past 2 years, it might make better economic and ergonomic sense to incorporate a dedicated microcontroller/DSP chip in a production apparatus. In its current configuration, the prototype image cytometer can be assembled, complete with laptop, for less than the price of most fluorescence microscopes. Figure 2 shows pictures of auramine O-stained positive and negative sputum slides taken by the prototype.

thumbnail image

Figure 2. Image cytometer pictures of positive and negative sputum slides stained with a low-background fast auramine O stain (74) (Scientific Device Laboratory).

Download figure to PowerPoint

The field of view in each of the 1× images in Figure 2 is ∼6.4 × 5.1 mm2. The bright spots in the positive smear that appear to be bacteria are colonies, but individual organisms can be seen in an enlarged image as single pixels with intensity above background level. Neither microcolonies nor individual organisms are visible in the image of the negative slide. We are confident that we can detect all acid-fast organisms in a smear; what remains unclear is whether or not an optimized staining technique can provide low enough nonspecific staining to allow the cytometer to be used for diagnosis on unprocessed sputum specimens collected in the field. We believe that the cytometer and an optimized stain will suffice to detect acid-fast organisms and microcolonies in smears of processed sputum, and that a clinical trial to test this hypothesis would be appropriate.

An image cytometer can easily detect signals from cells prepared as they would be for flow cytometry, while avoiding the aerosol generated by conventional flow cytometers; it is also easier for an image cytometer than for a flow cytometer to deal with MTB and other organisms that form aggregates. A colony-forming unit (CFU) may be an individual microbial cell or a microcolony containing more than one cell; cytometry can detect either, and effectively discriminate on the basis of size and DNA content. The use of a DNA-selective stain such as Pico Green allows image cytometry to determine two types of growth. In colonial growth, typical of MTB, the number of stained objects remains constant or nearly so, but the size and integrated fluorescence intensity (which reports DNA content) of these objects increases in proportion to the number of organisms per colony. Growth of single organisms, particularly motile ones, results in the appearance of more stained objects or events per field, but the size and intensity of objects remains relatively constant. We are, in essence, counting genomes in either case. The dye can be added to cultures in a formaldehyde solution [exposure to 1% paraformaldehyde for 40 min kills MTB (97)], making the samples safe for the cytometer operator to handle.

Although one can estimate cell numbers in microwells using a DNA-selective stain by measuring total fluorescence from individual wells, this requires high cell concentrations (typically 108–109/mL for bacteria, an impractical level to contemplate in the case of MTB), because background fluorescence from dye in solution masks fluorescence from small numbers of cells. With the image cytometer, it is feasible to use inocula of 100 or fewer organisms (or microcolonies)/well, and starting concentrations of 104 CFU/mL are more than adequate. The capacity of image cytometry to deal with low concentrations of organisms is advantageous because it shortens both the time required to obtain a starter culture and the incubation time required for susceptibility determination.

Figure 3 shows images of suspensions of Mycobacterium smegmatis, inoculated into wells of a 96-well microplate with and without drugs, and stained after 5 h (about two generation times) with 10 μM Pico Green (Molecular Probes/Invitrogen, Eugene, OR), a dye sufficiently DNA-selective to be routinely used for the quantification of DNA on gels. The small size of the reproduced images makes it difficult to see that there are many more bright spots in the untreated sample and the sample treated with 1 μg/mL INH, to which the organism is resistant, than in the sample treated with 5 μg/mL streptomycin, to which the organism is sensitive. We see no reason why MTB susceptibilities could not be determined within a similar number of generation times (24–48 h) using this methodology.

thumbnail image

Figure 3. M. smegmatis drug susceptibility by image cytometry.

Download figure to PowerPoint

THE MODS ASSAY AND CYTOMETRY

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED

The Microscopic Observation Direct Susceptibility (MODS) assay (90, 109, 110) for rapid TB diagnosis and susceptibility determination in resource-poor areas has recently attracted considerable attention. Decontaminated sputum is inoculated into multiwell plates containing liquid medium, with first-line drugs (typically INH and rifampin) in some wells, and the unstained wells are observed at relatively low power (40×) over a period of days. Growth of MTB is detected by the cordlike (“strings and tangles”) appearance of colonies. The MODS assay, which uses inexpensive consumables and simple equipment, determines antibiotic susceptibility at the same time as the diagnosis of TB is established [median 7 days in a comparative study (90)], and is therefore faster than current automated methodology, as well as considerably cheaper. It would be desirable to automate the readout of the MODS assay or develop a functional equivalent, in either case eliminating the need for a trained microscopist while yielding results in the same time frame.

Our cytometer can easily be converted to an inverted configuration, and the field of view enlarged to permit observation of individual wells of a 24- or 48-well plate instead of a 96-well plate, to make it completely compatible with the MODS assay format, although we have been advised that the assay can be done using the 96-well plates with which we already work.

If colonial growth is observed in a MODS culture before day 5, it is assumed that the culture has been contaminated by a rapidly growing organism, and it is discarded. By day 7, MTB colonies have typically increased their size a few hundred times, and colonial morphology allows them to be discriminated from other slow-growing organisms. There is, however, no reliable way in the context of the MODS assay to determine whether and how many acid-fast organisms are present in an average well at day 1. The cytometer and optimized auramine O stains allow us to do this. If acid-fast organisms are found in the input sample at day one, a fluorescent nucleic acid probe (87–89) could be used to determine whether they are MTB; determination of growth or lack thereof in the presence of INH and rifampin, which distinguishes MDR or XDR TB from more easily treatable disease, could be accomplished by day 4 or 5 using DNA stains.

We are also investigating the possibility that a simple cytochemical stain, based on very old observations correlating virulence, formation of cordlike colonies, and staining reactions (111, 112), and made tractable by recent work in molecular microbiology (113, 114) and neuroscience (115) could be used to discriminate virulent MTB from other slowly growing organisms. It may also be feasible to determine colonial morphology more directly at day 5 and beyond from low-resolution images of a MODS culture into which a relatively nontoxic green fluorescent redox indicator stain (RedoxSensor Green; Molecular Probes) is introduced at inoculation.

AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED

In their recent review, which established the advantages of fluorescence microscopy for TB diagnosis from sputum smears, Steingart et al. (34) concluded: “However, before changes in policy that support broad implementation of fluorescence microscopy can be considered, particularly in low-income countries, several issues need to be addressed: (1) feasibility and sustainability of fluorescence microscopy in settings with irregular electricity supply, limited human and financial resources, and inadequate training; (2) the lack of internationally agreed external quality assessment methods for blinded rechecking of fluorescent smears; (3) uncertainty about the stability of fluorescence microscopy reagents under field conditions; and (4) uncertainty about the acceptability of enclosed dark rooms to microscopists in tropical settings.” This sober assessment neither has stopped other workers in academia and industry from attempting to develop simpler, cheaper fluorescence microscopes, nor has it prevented their being funded for doing so.

Using the substage illuminator of our cytometer and its battery-operated power supply, and long pass or bandpass interference filters taped over the eyepieces, we have successfully adapted several binocular microscopes, the oldest dating from the 1930s, for fluorescence microscopy at a cost of less than U.S. $200 each. The power supply permits operation on batteries for hours at a time, but successful use of the instrument still requires a trained microscopist and a dark room and, as we have pointed out above, the relatively small volume of sample which a microscopist can practically examine still limits the accuracy and precision of both TB and malaria diagnosis. This also makes it difficult to recheck smears, because it is unlikely that successive observations of a slide will sample the same fields; the cytometer examines a larger field, and different instruments sample nearly identical areas.

The staining procedures we have suggested for use with the cytometer, i.e., an optimized, more specific auramine O stain for TB diagnosis and dyes that selectively stain DNA for malaria diagnosis and antimicrobial susceptibility testing of MTB, use reagents that should not be any less stable under field conditions than are Ziehl-Neelsen or Giemsa stains, and the cytometer requires neither a highly trained operator nor a dark room, is smaller, more robust, and less expensive than most fluorescence microscopes, and can be operated on batteries.

A fluorescence microscope itself cannot be used for the determination of antimicrobial susceptibility of MTB, but the cytometer can. The MODS test, which could provide susceptibility results in the same time frame as the cytometer, requires an inverted microscope, not standard equipment in laboratories in resource-poor areas, and a well-trained observer, but preparative techniques for the MODS and cytometric tests are similar in terms of their requirements for reagents and operator skills.

In affluent countries, microscopes are rarely used when cytometric alternatives are available. It appears to us that both the cost and complexity of cytometry have now decreased sufficiently for such alternatives to be made available to resource-poor countries as well.

FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED

Although we have not yet examined suitably stained slides of malaria parasites, the fact that we have had no problems detecting bacterial genomes about one-fifth the size of the P. falciparum genome and one-sixth the size of the P. vivax, P. malariae, and P. ovale genomes using our cytometer with the DNA-selective stains Pico Green and Vybrant DyeCycle Green (Molecular Probes) suggests to us that the apparatus should be able to detect all of the human malaria parasites. We are in the process of procuring specimens for further studies.

With slight modifications, notably a shift of excitation wavelength to ∼510 nm, and the use of either a color camera or two monochrome cameras with yellow and far red filters, we expect to be able to use the cytometer with phycoerythrin and phycoerythrin-Cy5 tandem labeled CD3 and CD4 antibodies for CD4+ T cell counting, and, if it is suitable for this application, it should be suitable for other immunophenotyping tasks. As Janossy et al. have noted elsewhere in this issue (116), diagnosis of TB in sputum-negative patients, particularly those with HIV/AIDS, may be facilitated by flow cytometric detection of specifically activated cells in blood (117); whether this job can be made simple enough for a minimalist imaging cytometer remains to be seen.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED

We believe, and have provided as much evidence as our own resources have to date permitted, that a small, simple, robust, inexpensive, energy-efficient low-resolution fluorescence image cytometer can be used in both resource-poor and developed countries for rapid, affordable diagnosis of malaria and TB and for rapid (24–48 h) determination of antimicrobial susceptibility of MTB. The apparatus is no more complex than a cellular phone and, indeed, utilizes components similar to those used in many such phones, i.e., a digital camera chip, a blue LED, and a microprocessor with wireless communications capability. Whether or not the instrument will meet the urgent needs that led to its conception remains to be determined; this will require a collaborative effort, in which we are more than eager to participate.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED

The authors thank Frank Mandy and George Janossy for first encouraging us to develop cytometric solutions to diagnostic problems in resource-poor areas, and those with whom we have also engaged in dialogue on this general subject: Tom Denny, Wolfgang Göhde, Sr., Peter Hansen, Jeff Harvey, Petra Krauledat, Chris Larsen, Helen Lee, Bala Manian, Coco Montagu, and Bill Rodriguez. Help with hardware, software, and reagents has come from Marco Angelini, Anne Carpenter, Bill Godfrey, Michael Stanley, Rob Webb, and Dane Wittrup. Our picture of malaria has been clarified by Jane Carlton, Martin Grobusch, Thomas Hänscheid, Jake Jacobberger, Michael Kariuki, Michael Makler, Anthony Moody, Steven Oh, Wendy Prudhomme O'Meara, Kyle Webster, and Innocent Yamodo. Our work on TB has benefited from interactions with Nuria Andreu, Koen Andries, Mike Barer, John Bernardo, Vanya Gant, Tom Garvey IV, Bob Gilman, Martin Grobusch, Thomas Hänscheid, Travis Hartman, Jim Huggett, Bill Jacobs, Gail Jacobs, Anil Koul, Stewart Lipton, Pat McGrath, Patrick Murray, Ed Nardell, Mark Perkins, Cidália Pina-Vaz, Lee Reichman, Ron Schell, Alan Schwartz, Alex Sloutsky, Henrik Stender, Armand van Deun, Torin Weisbrod, and Ying Zhang.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MALARIA, TB, AND CYTOMETRY: ANCIENT HISTORY AND MODERN BIOLOGY
  4. MALARIA AND TB DIAGNOSIS: BASIC CYTOMETRIC PROBLEMS
  5. DETERMINATION OF ANTIMICROBIAL SUSCEPTIBILITY IN TB
  6. A MINIMALIST FLUORESCENCE IMAGE CYTOMETER FOR RESOURCE-POOR AREAS
  7. THE MODS ASSAY AND CYTOMETRY
  8. AFFORDABLE FLUORESCENCE MICROSCOPY VS. AFFORDABLE CYTOMETRY
  9. FURTHER APPLICATIONS OF THE CYTOMETER: MALARIA AND BEYOND
  10. CONCLUSIONS
  11. Acknowledgements
  12. LITERATURE CITED
  • 1
    Gottlieb MS,Schroff R,Schanker HM,Weisman JD,Fan PT,Wolf RA,Saxon A. Pneumocystis carinii pneumonia and mucosal candidiasis in previously healthy homosexual men: Evidence of a new acquired cellular immunodeficiency. New Engl J Med 1981; 305: 14251431.
  • 2
    Fahey JL,Taylor JM,Detels R,Hofmann B,Melmed R,Nishanian P,Giorgi JV. The prognostic value of cellular and serologic markers in infection with human immunodeficiency virus type 1. N Engl J Med 1990; 322: 166172.
  • 3
    Mandy F,Nicholson J,Autran B,Janossy G. T-cell subset counting and the fight against AIDS: Reflections over a 20-year struggle. Cytometry 2002; 50: 3945.
  • 4
    Sharp PM,Bailes E,Chaudhuri RR,Rodenburg CM,Santiago MO,Hahn BH. The origins of acquired immune deficiency syndrome viruses: Where and when? Philos Trans R Soc Lond B Biol Sci 2001; 356: 867876.
  • 5
    Desowitz RS. The Malaria Capers. Tales of Parasites and People. New York: Norton; 1991. 288 p.
  • 6
    Ryan F. The Forgotten Plague: How the Battle Against Tuberculosis was Won—and Lost. Boston: Little, Brown and Company; 1993. xix + 460 pp.
  • 7
    Abu-Raddad LJ,Patnaik P,Kublin JG. Dual infection with HIV and malaria fuels the spread of both diseases in sub-Saharan Africa. Science 2006; 314: 16031606. Erratum in:Science2007;315:598.
  • 8
    Laufer MK,Plowe CV. The interaction between HIV and malaria in Africa. Curr Infect Dis Rep 2007; 9: 4754.
  • 9
    Friedland G,Churchyard GJ,Nardell E. Tuberculosis and HIV coinfection: Current state of knowledge and research priorities. J Infect Dis 2007; 196( Suppl 1): S1S3.
  • 10
    Wells CD,Cegielski JP,Nelson LJ,Laserson KF,Holtz TH,Finlay A,Castro KG,Weyer K. HIV infection and multidrug-resistant tuberculosis: The perfect storm. J Infect Dis 2007; 196( Suppl 1): S86S107.
  • 11
    Shapiro HM. “Cellular astronomy”—A foreseeable future in cytometry. Cytometry Part A 2004; 60A: 115124.
  • 12
    Shapiro HM,Perlmutter NG. Personal cytometers—Slow flow or no flow? Cytometry Part A 2009; 69A: 620630.
  • 13
    De Kruif P. Microbe Hunters (1926; republished). San Diego: Harcourt Brace; 1996. 357 p.
  • 14
    McNeill WH. Plagues and Peoples. New York: Doubleday; 1976. 340 p.
  • 15
    Gardner MJ,Hall N,Fung E,White O,Berriman M,Hyman RW,Carlton JM,Pain A,Nelson KE,Bowman S,Paulsen IT,James K,Eisen JA,Rutherford K,Salzberg SL,Craig A,Kyes S,Chan MS,Nene V,Shallom SJ,Suh B,Peterson J,Angiuoli S,Pertea M,Allen J,Selengut J,Haft D,Mather MW,Vaidya AB,Martin DM,Fairlamb AH,Fraunholz MJ,Roos DS,Ralph SA,McFadden GI,Cummings LM,Subramanian GM,Mungall C,Venter JC,Carucci DJ,Hoffman SL,Newbold C,Davis RW,Fraser CM,Barrell B. Genome sequence of the human malaria parasite Plasmodium falciparum. Nature 2002; 419: 498511.
  • 16
    McCutchan TF,Dame JB,Miller LH,Barnwell J. Evolutionary relatedness of Plasmodium species as determined by the structure of DNA. Science 1984; 225: 808811.
  • 17
    Carlton JM,Galinski MR,Barnwell JW,Dame JB. Karyotype and synteny among the chromosomes of all four species of human malaria parasite. Mol Biochem Parasitol 1999; 101: 2332.
  • 18
    Hammarton TC,Mottram JC,Doerig C. The cell cycle of parasitic protozoa: Potential for chemotherapeutic exploitation. Prog Cell Cycle Res 2003; 5: 91101.
  • 19
    Cole ST,Brosch R,Parkhill J,Garnier T,Churcher C,Harris D,Gordon SV,Eiglmeier K,Gas S,Barry CEIII,Tekaia F,Badcock K,Basham D,Brown D,Chillingworth T,Connor R,Davies R,Devlin K,Feltwell T,Gentles S,Hamlin N,Holroyd S,Hornsby T,Jagels K,Krogh A,McLean J,Moule S,Murphy L,Oliver K,Osborne J,Quail MA,Rajandream MA,Rogers J,Rutter S,Seeger K,Skelton J,Squares R,Squares S,Sulston JE,Taylor K,Whitehead S,Barrell BG. Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 1998; 393: 537544.
  • 20
    World Health Organization. Global tuberculosis control—Surveillance, planning, financing. WHO Report 2007 (WHO/HTM/TB/2007.376). Available at http://www.who.int/tb/publications/global_report/en.
  • 21
    Zignol M,Hosseini MS,Wright A,Weezenbeek CL,Nunn P,Watt CJ,Williams BG,Dye C. Global incidence of multidrug-resistant tuberculosis. J Infect Dis 2006; 194: 479485.
  • 22
    Centers for Disease Control and Prevention (CDC). Emergence of Mycobacterium tuberculosis with extensive resistance to second-line drugs worldwide, 2000–2004. MMWR Morb Mortal Wkly Rep 2006; 55: 301305.
  • 23
    Shah NS,Wright A,Bai GH,Barrera L,Boulahbal F,Martin-Casabona N,Drobniewski F,Gilpin C,Havelkova M,Lepe R,Lumb R,Metchock B,Portaels F,Rodrigues MF,Rusch-Gerdes S,Van Deun A,Vincent V,Laserson K,Wells C,Cegielski JP. Worldwide emergence of extensively drug-resistant tuberculosis. Emerg Infect Dis 2007; 13: 380387.
  • 24
    Parmet WE. Legal power and legal rights—Isolation and quarantine in the case of drug-resistant tuberculosis. N Engl J Med 2007; 357: 433435.
  • 25
    Hooke R. Micrographia. London: The Royal Society; 1665.Mineola, NY: Dover Publications;2003 (reprinted). 273 p.
  • 26
    Clark G,Kasten FH. History of Staining, 3rd ed. Baltimore: Williams & Wilkins; 1983. x + 304 p.
  • 27
    Garfield S. Mauve. New York: Norton; 2001. 222 p.
  • 28
    Stodola FH,Lesuk A,Anderson RJ. The chemistry of the lipids of tubercle bacilli. LIV. The isolation and properties of mycolic acid. J Biol Chem 1938; 126: 505513.
  • 29
    Asselineau J,Lederer E. Structure of the mycolic acids of mycobacteria. Nature 1950; 166: 782783.
  • 30
    Barry CEIII,Lee RE,Mdluli K,Sampson AE,Schroeder BG,Slayden RA,Yuan Y. Mycolic acids: Structure, biosynthesis and physiological functions. Prog Lipid Res 1998; 37: 143179.
  • 31
    Watanabe M,Aoyagi Y,Ridell M,Minnikin DE. Separation and characterization of individual mycolic acids in representative mycobacteria. Microbiology 2001; 147: 18251837.
  • 32
    Takayama K,Wang C,Besra GS. Pathway to synthesis and processing of mycolic acids in Mycobacterium tuberculosis. Clin Microbiol Rev 2005; 18: 81101.
  • 33
    Hagemann PKH. Fluoreszenzfärbung von Tuberkelbakterien mit Auramin. Münch Med Wschr 1938; 85: 10661068.
  • 34
    Steingart KR,Henry M,Ng V,Hopewell PC,Ramsay A,Cunningham J,Urbanczik R,Perkins M,Aziz MA,Pai M. Fluorescence versus conventional sputum smear microscopy for tuberculosis: A systematic review. Lancet Infect Dis 2006; 6: 570581.
  • 35
    Anthony RM,Kolk AH,Kuijper S,Klatser PR. Light emitting diodes for auramine O fluorescence microscopic screening of Mycobacterium tuberculosis. Int J Tuberc Lung Dis 2006; 10: 10601062.
  • 36
    Hänscheid T,Ribeiro CM,Shapiro HM,Perlmutter NG. Fluorescence microscopy for tuberculosis diagnosis. Lancet Infect Dis 2007; 7: 236237.
  • 37
    Hung NV,Sy DN,Anthony RM,Cobelens FG,van Soolingen D. Fluorescence microscopy for tuberculosis diagnosis. Lancet Infect Dis 2007; 7: 238239.
  • 38
    Fleischer B. 100 years ago: Giemsa's solution for staining of plasmodia. Trop Med Int Health 2004; 9: 755756.
  • 39
    Ambroise-Thomas P,Michel-Brun J,Despeignes J. [Rapid identification of sanguicolous parasites by staining with acridine orange and fluorescence microscopy]. Bull Soc Pathol Exot Filiales 1965; 58: 639643.
  • 40
    Shute GT,Sodeman TM. Identification of malaria parasites by fluorescence microscopy and acridine orange staining. Bull World Health Organ 1973; 48: 591596.
  • 41
    Spielman A,Perrone JB,Teklehaimanot A,Balcha F,Wardlaw SC,Levine RA. Malaria diagnosis by direct observation of centrifuged samples of blood. Am J Trop Med Hyg 1988; 39: 337342.
  • 42
    Makler MT,Ries LK,Ries J,Horton RJ,Hinrichs DJ. Detection of Plasmodium falciparum infection with the fluorescent dye, benzothiocarboxypurine. Am J Trop Med Hyg 1991; 44: 1116.
  • 43
    Malinin GI,Malinin TI. Rapid microscopic detection of malaria parasites permanently fluorochrome stained in blood smears with aluminum and morin. Am J Clin Pathol 1991; 95: 424427.
  • 44
    Caramello P,Lucchini A,Savoia D,Gioannini P. Rapid diagnosis of malaria by use of fluorescent probes. Diagn Microbiol Infect Dis 1993; 17: 293297.
  • 45
    Ochola LB,Vounatsou P,Smith T,Mabaso ML,Newton CR. The reliability of diagnostic techniques in the diagnosis and management of malaria in the absence of a gold standard. Lancet Infect Dis. 2006; 6: 582588.
  • 46
    “Student” [ Gosset WS]. On the error of counting with a haemacytometer. Biometrika 1907; 5: 351360.
  • 47
    Moody A. Rapid diagnostic tests for malaria parasites. Clin Microbiol Rev 2002; 15: 6678.
  • 48
    Hänscheid T. Diagnosis of malaria: A review of alternatives to conventional microscopy. Clin Lab Haematol 1999; 21: 235245.
  • 49
    O'Meara WP,McKenzie FE,Magill AJ,Forney JR,Permpanich B,Lucas C,Gasser RAJr,Wongsrichanalai C. Sources of variability in determining malaria parasite density by microscopy. Am J Trop Med Hyg 2005; 73: 593598.
  • 50
    O'Meara WP,Hall BF,McKenzie FE. Malaria vaccine efficacy: The difficulty of detecting and diagnosing malaria. Malaria J 2007; 6: 36.
  • 51
    Wittekind DH. On the nature of Romanowsky-Giemsa staining and its significance for cytochemistry and histochemistry: An overall view. Histochem J 1983; 15: 10291047.
  • 52
    Zipfel E,Grezes JR,Naujok A,Seiffert W,Wittekind DH,Zimmermann HW. [Romanowsky dyes and the Romanowsky-Giemsa effect. 3. Microspectrophotometric studies of Romanowsky-Giemsa staining. Spectroscopic evidence of a DNA-azure B-eosin Y complex producing the Romanowsky-Giemsa effect.] Histochemistry 1984; 81: 337351 (in German).
  • 53
    Horobin RW,Walter KJ. Understanding Romanowsky staining. I. The Romanowsky-Giemsa effect in blood smears. Histochemistry 1987; 86: 331336.
  • 54
    Shapiro HM,Mandy F. Cytometry in malaria: Moving beyond Giemsa. Cytometry Part A 2007; 71A: 643645.
  • 55
    Jackson PR,Winkler DG,Kimzey SL,Fisher FM. Cytofluorograf detection of Plasmodium yoelii, Trypanosoma gambiense, and Trypanosoma equiperdum by laser excited fluorescence of stained rodent blood. J Parasitol 1977; 63: 593598.
  • 56
    Howard RJ,Battye FL,Mitchell GF. Plasmodium-infected blood cells analyzed and sorted by flow fluorimetry with the deoxyribonucleic acid binding dye 33258 Hoechst. J Histochem Cytochem 1979; 27: 803813.
  • 57
    Xie L,Li Q,Johnson J,Zhang J,Milhous W,Kyle D. Development and validation of flow cytometric measurement for parasitaemia using autofluorescence and YOYO-1 in rodent malaria. Parasitology 2007; 134: 11511162.
  • 58
    Bhakdi SC,Sratongno P,Chimma P,Rungruang T,Chuncharunee A,Neumann HP,Malasit P,Pattanapanyasat K. Re-evaluating acridine orange for rapid flow cytometric enumeration of parasitemia in malaria-infected rodents. Cytometry Part A 2007; 71A: 662667.
  • 59
    Hare JD,Bahler DW. Analysis of Plasmodium falciparum growth in culture using acridine orange and flow cytometry. J Histochem Cytochem 1986; 34: 215220.
  • 60
    Hare JD. Two-color flow-cytometric analysis of the growth cycle of Plasmodium falciparum in vitro: Identification of cell cycle compartments. J Histochem Cytochem 1986; 34: 16511658.
  • 61
    Hänscheid T,Egan TJ,Grobusch MP. Haemozoin: From melatonin pigment to drug target, diagnostic tool, and immune modulator. Lancet Infect Dis 2007; 7: 675685.
  • 62
    Mendelow BV,Lyons C,Nhlangothi P,Tana M,Munster M,Wypkema E,Liebowitz L,Marshall L,Scott S,Coetzer TL. Automated malaria detection by depolarization of laser light. Br J Haematol 1999; 104: 499503.
  • 63
    Krämer B,Grobusch MP,Suttorp N,Neukammer J,Rinneberg H. Relative frequency of malaria pigment-carrying monocytes of nonimmune and semi-immune patients from flow cytometric depolarized side scatter. Cytometry 2001; 45: 133140.
  • 64
    Grobusch MP,Hänscheid T,Krämer B,Neukammer J,May J,Seybold J,Kun JF,Suttorp N. Sensitivity of hemozoin detection by automated flow cytometry in non- and semi-immune malaria patients. Cytometry Part B Clin Cytom 2003; 55B: 4651.
  • 65
    Steingart KR,Ng V,Henry M,Hopewell PC,Ramsay A,Cunningham J,Urbanczik R,Perkins MD,Aziz MA,Pai M. Sputum processing methods to improve the sensitivity of smear microscopy for tuberculosis: A systematic review. Lancet Infect Dis 2006; 6: 664674.
  • 66
    MurrayPR, BaronEJ, JorgensenJH, LandryML, PfallerMA, editors. Manual of Clinical Microbiology, 9th ed. Washington: ASM Press; 2007. 2488 p.
  • 67
    Richards OW. The staining of acid-fast tubercle bacteria. Science 1941; 93: 190.
  • 68
    Oster G. Fluorescence of auramine O in the presence of nucleic acid. C R Hebd Seances Acad Sci 1951; 232: 17081710 (in French).
  • 69
    Kojima K,Niri M,Setoguchi K,Tsuda I,Tatsumi N. An automated optoelectronic reticulocyte counter. Am J Clin Pathol 1989; 92: 5761.
  • 70
    Ibrahim P,Whiteley AS,Barer MR. SYTO 16 labelling and flow cytometry of Mycobacterium avium. Lett Appl Microbiol 1997; 25: 437441.
  • 71
    Pina-Vaz C,Costa-de-Oliveira S,Rodrigues AG. Safe susceptibility testing of Mycobacterium tuberculosis by flow cytometry with the fluorescent nucleic acid stain SYTO 16. J Med Microbiol 2005; 54: 7781.
  • 72
    Pina-Vaz C,Sofia Costa-Oliveira S,Goncalves Rodrigues A,Salvador A. Novel method using a laser scanning cytometer for detection of mycobacteria in clinical samples. J Clin Microbiol 2004; 42: 906908.
  • 73
    Goren MB,Cernich M,Brokl O. Some observations of mycobacterial acid-fastness. Am Rev Respir Dis 1978; 118: 151154.
  • 74
    Draper P. The outer parts of the mycobacterial envelope as permeability barriers. Front Biosci 1998; 3: D1253D1261.
  • 75
    Schlacks M,Coppernoll S,DeBoo A,Schmidt A,Bartnicki L,Schreckenberger P,Lipton S. Evaluation of four commercially available auramine O stain sets. Presented at the American Society for Microbiology General Meeting, Orlando, FL, May 2006. Poster C-324.
  • 76
    Veropoulos K,Learmonth G,Campbell C,Knight B,Simpson J. Automated identification of tubercle bacilli in sputum. A preliminary investigation. Anal Quant Cytol Histol 1999; 21: 277282.
  • 77
    Forero MG,Cristóbal G,Desco M. Automatic identificationof Mycobacterium tuberculosis by Gaussian mixture models. J Microsc 2006; 223(Part 2): 120132.
  • 78
    Getahun H,Harrington M,O'Brien R,Nunn P. Diagnosis of smear-negative pulmonary tuberculosis in people with HIV infection or AIDS in resource-constrained settings: Informing urgent policy changes. Lancet 2007; 369: 20422049.
  • 79
    Colebunders R,Bastian A. A review of the diagnosis and treatment of smear-negative pulmonary tuberculosis. Int J Tuberc Lung Dis 2000; 4: 97107.
  • 80
    Rouillon A,Perdrizet S,Parrot R. Transmission of tubercle bacilli: The effects of chemotherapy. Tubercle 1976; 57: 275299.
  • 81
    FriedenT, editor. Toman's Tuberculosis. Case Detection, Treatment, and Monitoring—Questions and Answers,2nd ed. Geneva: World Health Organization; 2004. pp 1121. Available at http://www.who.int/tb/publications/toman/en/index.html (online free).
  • 82
    Hobby GL,Holman AP,Iseman MD,Jones JM. Enumeration of tubercle bacilli in sputum of patients with pulmonary tuberculosis. Antimicrob Agents Chemother 1973; 4: 94104.
  • 83
    Chakravorty S,Tyagi JS. Novel multipurpose methodology for detection of mycobacteria in pulmonary and extrapulmonary specimens by smear microscopy, culture, and PCR. J Clin Microbiol 2005; 43: 26972702.
  • 84
    Haldar S,Chakravorty S,Bhalla M,De Majumdar S,Tyagi JS. Simplified detection of Mycobacterium tuberculosis in sputum using smear microscopy and PCR with molecular beacons. J Med Microbiol 2007; 56: 13561362.
  • 85
    Ozanne V,Ortalo-Magne A,Vercellone A,Fournie JJ,Daffe M. Cytometric detection of mycobacterial surface antigens: Exposure of mannosyl epitopes and of the arabinan segment of arabinomannans. J Bacteriol 1996; 178: 72547259.
  • 86
    Nader WF,Nebe CT,Nebe G,Dastani A,Birr C. Analysis of bacteria in environmental and medical microbiology by flow cytometry. In: VeheriA, TiltonRC, BalowsA, editors. Rapid Methods and Automation in Microbiology and Immunology. Berlin: Springer; 1991. pp 131140.
  • 87
    Stender H,Mollerup TA,Lund K,Petersen KH,Hongmanee P,Godtfredsen SE. Direct detection and identification of Mycobacterium tuberculosis in smear-positive sputum samples by fluorescence in situ hybridization (FISH) using peptide nucleic acid (PNA) probes. Int J Tuberc Lung Dis 1999; 3: 830837.
  • 88
    Stender H,Lund K,Petersen KH,Rasmussen OF,Hongmanee P,Miorner H,Godtfredsen SE. Fluorescence in situ hybridization assay using peptide nucleic acid probes for differentiation between tuberculous and nontuberculous mycobacterium species in smears of mycobacterium cultures. J Clin Microbiol 1999; 37: 27602765.
  • 89
    Hongmanee P,Stender H,Rasmussen OF. Evaluation of a fluorescence in situ hybridization assay for differentiation between tuberculous and nontuberculous Mycobacterium species in smears of Lowenstein-Jensen and Mycobacteria growth indicator tube cultures using peptide nucleic acid probes. J Clin Microbiol 2001; 39: 10321035.
  • 90
    Moore DA,Evans CA,Gilman RH,Caviedes L,Coronel J,Vivar A,Sanchez E,Pinedo Y,Saravia JC,Salazar C,Oberhelman R,Hollm-Delgado MG,LaChira D,Escombe AR,Friedland JS. Microscopic-observation drug-susceptibility assay for the diagnosis of TB. N Engl J Med 2006; 355: 15391550.
  • 91
    Resnick M,Schuldiner S,Bercovier H. Bacterial membrane potential analyzed by spectrofluorocytometry. Curr Microbiol 1985; 12: 183186.
  • 92
    Novo DJ,Perlmutter NG,Hunt RH,Shapiro HM. Multiparameter flow cytometric analysis of antibiotic effects on membrane potential, membrane permeability, and bacterial counts of Staphylococcus aureus and Micrococcus luteus. Antimicrob Agents Chemother 2000; 44: 827834.
  • 93
    Silverman JA,Perlmutter NG,Shapiro HM. Correlation of daptomycin bactericidal activity and membrane depolarization in Staphylococcus aureus. Antimicrob Agents Chemother 2003; 47: 25382544.
  • 94
    Norden MA,Kurzynski TA,Bownds SE,Callister SM,Schell RF. Rapid susceptibility testing of Mycobacterium tuberculosis (H37Ra) by flow cytometry. J Clin Microbiol 1995; 33: 12311237.
  • 95
    Bownds SE,Kurzynski TA,Norden MA,Dufek JL,Schell RF. Rapid susceptibility testing for nontuberculosis mycobacteria using flow cytometry. J Clin Microbiol 1996; 34: 13861390.
  • 96
    Kirk SM,Schell RF,Moore AV,Callister SM,Mazurek GH. Flow cytometric testing of susceptibilities of Mycobacterium tuberculosis isolates to ethambutol, isoniazid, and rifampin in 24 hours. J Clin Microbiol 1998; 36: 15681573.
  • 97
    Moore AV,Kirk SM,Callister SM,Mazurek GH,Schell RF. Safe determination of susceptibility of Mycobacterium tuberculosis to antimycobacterial agents by flow cytometry. J Clin Microbiol 1999; 37: 479483.
  • 98
    Vena RM,Munson EL,DeCoster DJ,Croke CL,Fett DB,Callister SM,Schell RF. Flow cytometric testing of susceptibilities of Mycobacterium avium to amikacin, ciprofloxacin, clarithromycin and rifabutin in 24 hours. Clin Microbiol Infect 2000; 6: 368375.
  • 99
    Schell RF,Nardelli DT,DeCoster DJ,Kirk SM,Callister SM. Mycobacterium tuberculosis susceptibility testing by flow cytometry. Unit 11.7. In: RobinsonJP, DarzynkiewiczZ, DeanP,HibbsAR, OrfaoA, RabinovitchP, WheelessL, editors. Current Protocols in Cytometry. New York: Wiley; 2004. pp 11.7.111.7.8.
  • 100
    DeCoster DJ,Vena RM,Callister SM,Schell RF. Susceptibility testing of Mycobacterium tuberculosis: Comparison of the BACTEC TB-460 method and flow cytometric assay with the proportion method. Clin Microbiol Infect 2005; 11: 372378.
  • 101
    Fredricks BA,Decoster DJ,Kim Y,Sparks N,Callister SM,Schell RF. Rapid pyrazinamide susceptibility testing of Mycobacterium tuberculosis by flow cytometry. J Microbiol Methods 2006; 67: 266272.
  • 102
    Ryan C,Nguyen BT,Sullivan SJ. Rapid assay for mycobacterial growth and antibiotic susceptibility using gel microdrop encapsulation. J Clin Microbiol 1995; 33: 17201726.
  • 103
    Reis RS,Neves IJr,Lourenco SL,Fonseca LS,Lourenco MC. Comparison of flow cytometric and Alamar Blue tests with the proportional method for testing susceptibility of Mycobacterium tuberculosis to rifampin and isoniazid. J Clin Microbiol 2004; 42: 22472248.
  • 104
    Akselband Y,Cabral C,Shapiro DS,McGrath P. Rapid mycobacteria drug susceptibility testing using Gel Microdrop (GMD) Growth Assay and flow cytometry. J Microbiol Methods 2005; 62: 181197.
  • 105
    Stein PG,Lipkin LE,Shapiro HM. Spectre II. General-purpose microscope input for a computer. Science 1969; 166: 328333.
  • 106
    Shapiro HM,Bryan SD,Lipkin LE,Stein PG,Lemkin PF. Computer-aided microspectrophotometry of biological specimens. Exp Cell Res 1971; 67: 8189.
  • 107
    Mazzini G,Ferrari C,Baraldo N,Mazzini M,Angelini M. Improvements in fluorescence microscopy allowed by high power light emitting diodes. In: Méndez-VilasA, Labajos-BroncanoL, editors. Current Issues on Multidisciplinary Microscopy Research and Education, Vol. 2. Badajoz, Spain: FORMATEX; 2005. pp 181188. Available at http://www.formatex.org/micro2003/papers/181–188. pdf.
  • 108
    Wittrup KD,Westerman RJ,Desai R. Fluorescence array detector for large-field quantitative fluorescence cytometry. Cytometry 1994; 16: 206213.
  • 109
    Caviedes L,Lee TS,Gilman RH,Sheen P,Spellman E,Lee EH,Berg DE,Montenegro-James S. Rapid, efficient detection and drug susceptibility testing of Mycobacterium tuberculosis in sputum by microscopic observation of broth cultures. J Clin Microbiol 2000; 38: 12031208.
  • 110
    Caviedes L,Coronel J,Evans C,Leonard B,Gilman R,Moore D. MODS—A user guide. 2006. Available at http://www.upch.edu.pe/facien/dbmbqf/MODS_user_guide.pdf.
  • 111
    Middlebrook G,Dubos RJ,Pierce C. Virulence and morphologic characteristics of mammalian tubercle bacilli. J Exp Med 1947; 86: 175184.
  • 112
    Dubos RJ,Middlebrook G. Cytochemical reaction of virulent tubercle bacilli. Am Rev Tuberc 1948; 58: 698699.
  • 113
    Andreu N,Soto CY,Roca I,Martin C,Gibert I. Mycobacterium smegmatis displays the Mycobacterium tuberculosis virulence-related neutral red character when expressing the Rv0577 gene. FEMS Microbiol Lett 2004; 231: 283289.
  • 114
    Cardona PJ,Soto CY,Martin C,Giquel B,Agusti G,Guirado E,Sirakova T,Kolattukudy P,Julian E,Luquin M. Neutral-red reaction is related to virulence and cell wall methyl-branched lipids in Mycobacterium tuberculosis. Microbes Infect 2006; 8: 183190.
  • 115
    Okada D. Neutral red as a hydrophobic probe for monitoring neuronal activity. J Neurosci Methods 2000; 101: 8592.
  • 116
    Janossy G,Barry SM,Breen RAM,Hardy G,Lipman M,Kern F. The role of clinical flow cytometry in the interferon-gamma based diagnosis of active tuberculosis and its co-infection with HIV-1—A technically oriented review. Cytometry Part B Clin Cytom 2008; 74B ( Suppl): in press. doi://10.1002/cyto.b.20381 (this issue).
  • 117
    Streitz M,Tesfa L,Yildirim V,Yahyazadeh A,Ulrichs T,Lenkei R,Quassem A,Liebetrau G,Nomura L,Maecker H,Volk HD,Kern F. Loss of receptor on tuberculin-reactive T-cells marks active pulmonary tuberculosis. PLoS ONE 2007; 2: e735.