SEARCH

SEARCH BY CITATION

Keywords:

  • spectroscopic characterization;
  • microorganisms;
  • Fourier Transform Infrared Microspectroscopy

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSION
  7. REFERENCES

Spectroscopic fingerprints of bacteria were investigated by Fourier transform infrared (FTIR) microspectroscopy for the elucidation of chemical composition and structural information during growth. Good differentiation of six microorganisms was achieved down to the strain level. The inherent compositional and structural differences of cell envelopes and cytoplasm were investigated and utilized to obtain more detailed analysis of the spectroscopic features. Bands or regions of key functional groups were also identified in the original spectra. Microspectroscopic monitoring of bacterial growth demonstrated that FTIR spectroscopy cannot only provide molecular fingerprints of the cell envelope, but also compositional and metabolic information of the cytoplasm under different physiological conditions. This approach could be an effective alternative to traditional nutritional and biochemical methods to monitor and assess the effects of inhibitors and other environmental factors on microbial cell growth. © 2005 Wiley Periodicals, Inc. Biopolymers 77: 368–377, 2005


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSION
  7. REFERENCES

Fourier transforms infrared spectroscopy (FTIR), a widely used technique in biophysics and biochemistry, provides structural information of biological molecules such as proteins, nucleic acids, carbohydrates, and lipids. In the past decade, FTIR has also been applied to the identification and differentiation of microorganisms. Since Naumann and coworkers published their pioneering work in this field,1 various research groups around the world have shown the validity of FTIR spectroscopic techniques beyond reasonable doubt.2–7 Results obtained have consistently proved that the spectral information is sufficient to distinguish between various microorganisms both at species and strain level. In particular, IR microspectrometry can be used to provide spatially resolved structural and compositional information at the molecular level, accounting for the heterogeneities of the microbial colonies at the microscopic level for enhanced discrimination. Objective structural information of the microscopic structures could be obtained from the spatially resolved IR mapping of the sample areas.8 These maps reflect the spatial distribution of functional groups (chemical mapping) or of the specific IR spectroscopic patterns (pattern mapping), and can be directly compared with the histological specimens produced by conventional staining procedures.9

Although it is generally accepted that IR spectra of biological materials provide characteristic information of the chemical composition and structure, because of the overlapping absorbance due to multitude of cellular compounds, the bands observed in the mid-IR range are not highly resolved and hence it is very difficult to gain a comprehensive understanding of the biomolecules from the spectral information. A common strategy is to use multivariate data analysis techniques such as factor analysis, cluster analysis, or artificial neural network analysis to study the general pattern hidden in the microbial spectral data to differentiate, identify, and classify microorganisms without a priori information.10 Although considerable advances have been made in applying FTIR microspectroscopy for microorganism identification and differentiation following this approach, few have focused on examining the chemical and structural differences that are the real basis of these different patterns. Several efforts have been made to assign specific bands to functional groups that are biologically important,6, 11 but they are incomplete and some of the assignments proposed are still debatable. Most of these efforts are based on the first or second derivatives of the original (zero-order) spectrum. Although derivative spectra can enhance the resolution by revealing the finer structures usually observed as broad peaks in the zero-order spectra, they also amplify the noise signals. In order to avoid deterioration of the signal-to-noise (S/N) ratio in the derivative spectra, smoothing is almost inevitable, but noting that information might be lost due to oversmoothing. The signal intensity in derivative spectra is interrelated to the full width at half height (FWHH) of the respective peaks in the zero-order spectra, but does not reflect the absorbance intensity directly.17 For quantitative analysis of the spectral data, calibration using known material is required. In spectroscopic study of microorganisms, calibration with a known material is almost impossible—hence using derivative spectra for quantitative analysis must be conducted with caution. FTIR microspectrometry when properly employed might provide highly resolved zero-order spectra for cellular observations and the interpretation of the spectral data is easier—and hence may facilitate band assignments on the zero-order spectrum.

The composition of the cell envelopes (the cytoplasmic membrane and the cell wall) and the cytoplasm are quite different; lipid and polysaccharides are the main building bricks of the cell envelopes, on the other hand, proteins and nucleic acid dominate the cytoplasm. By comparing the spectral features of the whole cell to the spectra of the cell envelopes and cytoplasm separately, specific details of the spectral features can be assessed. From the contributions of each of the main cell constituents, compositional and structural features that can serve as molecular indicators can be identified and used as markers for cell growth and characterization.

The objective of this study is to apply FTIR microspectroscopy to whole microbial cells as well as their separated envelopes and cytoplasm to establish a spectroscopic fingerprinting basis for effective discrimination of microbial cells. In this study we will also continuously monitor the spectroscopic changes during culturing of a bacterium to establish a relationship between these changes and traditional growth curve to serve as a first step to monitoring compositional and structural changes during microbial cell development by IR spectroscopy.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSION
  7. REFERENCES

Sample Preparation

Pathogenic strains of Salmonella (enteritidis and typhimurium), Escherichia coli (serotype O26, O27 and O157:H7), Yersinia enterocolitis, and Shigella boydii were obtained from the Gastroenteric Disease Center (GDC) at the Pennsylvania State University (University Park, PA). Microspectroscopic measurements were performed on all of these microorganisms to obtain IR spectra of whole cells. The cytoplasm and cell envelopes of four species (S. enteritidis, S. Typhimurium, E. coli O27 and O157:H7) were separated and examined for specific markers for differentiation and characterization.

These species were cultured on soy agar plates for 24 h, and a single colony of each species was transferred into 100 mL broth medium (5 g yeast extract, 8 g tryptone, and 5 g NaCl in 500 mL distilled water) at 37°C and shaken at 100 rpm for 24 h. Ten millilters of each culture was centrifuged at 3000 rpm for 25 min and the microbial cells were collected. Then they were washed three times by distilled water to remove residual medium and to ready them for microspectroscopic measurements.

For the separation of the cell envelopes and cytoplasmic extract, the collected cells were washed 3 times and resuspended in 5 mL distilled water, and the pH was adjusted to 12 by slowly adding 0.1M NaOH. Escherichia coli and Salmonella cells were lysed at this pH.12 The suspension was then centrifuged at 13,000 rpm for 45 min; the cell envelope was collected and its IR spectra were measured by microspectrometry. The supernatant contains the cytoplasmic extract, which is water soluble, and its IR spectra were measured by a regular FTIR spectrometer using an attenuated total reflectance (ATR) setting.

For monitoring the bacterial growth and changes in the spectroscopic features, S. boydii was investigated. The bacterium was cultured on soy agar plates for 24 h, and a single colony was transferred into 100 mL broth medium (5 g yeast extract, 8 g tryptone, and 5 g NaCl in 500 mL distilled water) at 37°C and shaken at 100 rpm. A 5 mL sample was taken after 3, 6, 9, 12, 24, 36, and 48 h. The optical density of the samples was measured by a UV/Visible spectrophotometer (DU 530, Beckman Coulter, Inc., Fullerton, CA, USA) and a growth curve was constructed. The cells were periodically collected and prepared for microspectroscopic measurements. The cells were lysed to isolate the cytoplasmic extract from the cell envelope and their spectra were measured by a regular FTIR spectrometer using the ATR setting.

FTIR Measurements Using ATR

The IR spectra of the cytoplasmic extract, which is a suspension in distilled water, were measured using a Digilab Excalibur FTS 3000 spectrometer (Digilab, Randolph, MA) with a ceramic air-cooled source, a room-temperature-operated deuterated triglycine sulfate detector (DTGS) detector, a KBr beamsplitter, and a ZnSe ATR accessory (8 bounce). The angle of the incidence of the IR beam was 45°. Each time an 80 μL of sample suspension was loaded onto the ATR crystal and FTIR signals were collected in the spectral region between 600 and 4000 cm−1, at a resolution of 2 cm−1 at room temperature. Distilled water was measured first and used as the background, and ratioed out from the sample spectra. To obtain the spectra of each sample, 256 scans were averaged. The intensity differences caused by wave number dependence of penetration depth were automatically corrected by the software (Merlin 3.4, Digilab, Randolph, MA). Twelve spectra were taken for each bacterial species. As shown in Figure 1, only regions between 800–1800 cm−1 and 2700 an 3100 cm−1, which are the “fingerprints” region for microorganism differentiation, were used for analysis.6, 7

thumbnail image

Figure 1. Spectra of different microorganisms measured by FTIR microspectrometry.

Download figure to PowerPoint

FTIR Microspectrometry

Whole cell samples and cell envelope samples were smeared onto a gold-coated glass slide (∼200-nm thick gold layer) by an aseptic swab to form a ∼20-mm2 spot and then dried 24 h in a dessicator for microspectroscopic measurements. The mid-IR spectra were measured using a Digilab Excalibur FTS 6000 spectrometer fitted with a UMA 600 IR microscope (Digilab, Randolph, MA, USA) with a liquid-nitrogen-cooled mercury-cadmium-telluride (MCT) detector. IR spectrum from a 200 × 200 μm sample area of the sample spot (∼20 mm2) on the gold slide could be obtained using the maximum aperture. The sample chamber was purged with helium gas each time before a measurement was taken to minimize interference from water vapor. Twelve spectra were measured for each sample spot at different locations, in the spectral region between 600 and 4000 cm−1, at a resolution of 8 cm−1, with 256 scans averaged. The IR source was a ceramic air-cooled source, and the beam used was KBr. Experiments were performed at room temperature and the regions between 800–1800 cm−1 and 2700–3100 cm−1 were used for analysis.

Data Normalization

All spectra measured were normalized against the amide I band, which was used as an internal standard.4 For the same microorganism, under identical growing conditions, the relative intensity of each peak against the internal standard for cells at the same growing stage should be constant; effects of baseline shift and other variations due to instrument fluctuation were removed. For cells at different growing stages, because of the changes in the internal chemical composition, the normalized spectra would show different characteristics and could be used in a discriminant analysis to differentiate between growing stages.

For reproducibility of the data, it was important to establish a strict experimental protocol relative to media, incubation time, temperature, harvest of bacteria, and sample preparation. Therefore, these experimental conditions as well as the spectroscopic measurement settings were repeated for each FTIR measurements.

Discriminant Analysis

FTIR fingerprints thus obtained from ATR and microscope measurements were further processed by chemometrics. The spectral data were first subjected to principal component analysis (PCA) to reduce the dimension of the data set. Then canonical variate analysis (CVA) was conducted on the reduced data set in which weighted linear combinations were structured to maximize the difference among the group means, relative to their variances. The linear combinations of the variances, the variates, were then plotted and analyzed by cluster analysis, each microorganism, or each growing stage was represented by one cluster. Both PCA and CVA were conducted using Win-DAS (Wiley & Sons, Chichester, UK, 1998 version).

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSION
  7. REFERENCES

Microspectroscopic Characterization of Microorganisms

Figure 1 shows the IR spectra of whole cell samples of six microorganisms measured by microspectrometry in the wavenumber ranges between 800–1800 cm−1 and 2700–3100 cm−1. Possibly because of the enhancement effect of the gold surface, contributions from the DNA of the microorganisms11 can be observed in these spectra compared to traditional ATR results, especially in the 800–900-cm−1 range. Furthermore, the amide I bands (1637–1695 cm−1) characteristic of the secondary structure of proteins, shown in Figure 1b, show species-specific patterns. In the range of 2700-–100 cm−1, features that represent fatty acid and lipid contributions could be observed. Since the spectra were normalized against the amide I band, the difference in the absorbance intensity show that the overall concentration of fatty acids and lipids vary with respect to the chemical composition of these microorganisms and therefore can serve as markers for differentiation. Figure 2 shows the differentiation by chemometric analysis of the microspectroscopic data. It can be seen that the three E. coli strains form closely clustered supergroups and the S. boydii group shows a closer similarity to these E. coli strains than S. typhimurium and Y. enterocolitis, demonstrating the taxonomic relations between these bacteria. When the data of the three E. coli strains were analyzed independently, they could be very well differentiated (Figure 2b). The loading numbers for the first 2 PCs are shown in Figure 2c. It is clear that, based on microspectroscopic data, not only microorganisms can be differentiated, but also their taxonomic relationships deduced.

thumbnail image

Figure 2. Differentiation of microorganisms by chemometrics analysis. (a) All microbial species; (b) three E. coli strains; (c) loading numbers for the first 2 PCs.

Download figure to PowerPoint

Spectroscopic Characterization of Cytoplasmic Extract and Cell Envelopes

The IR spectra measured for microbial cells are a superposition of contributions from all biomolecules present in a cell; therefore, the peaks observed are broad and hence the specific contribution from any particular biomolecules or molecular groups are difficult to assess. Since the chemical composition of the cell envelope and cytoplasmic extract have major differences, the basic framework of the cytoplasmic membrane are due to the phosphorous lipid bilayers while the main framework of the cell wall is made of polysaccharides, and the proteins present in them often have polysaccharide side chains attached to it. Therefore, we can expect the polysaccharide and lipid contributions in the spectra of cell envelopes to be prominent. As we know, all antibody binding sites and other signal receptors are located at the surface of the cell envelopes; thus the differences recognized in immunoassay techniques that serve as the basis for differentiation of microorganisms are detectable in the spectroscopic features of the cell envelope.

On the other hand, RNA/DNA is one of the major constituents of the cytoplasmic extract, especially for the cells that divide rapidly. A large number of carbohydrates are also present in the cytoplasm. We expect the amide I and II bands to be less prominent in this spectroscopic landscape. Also, the chemical composition of the cytoplasm for the same species can vary during the development, or under different growing conditions. Therefore the spectroscopic feature of the cytoplasm may not be a good tool to differentiate different microorganisms; however, it should be a useful tool to investigate changes induced by different physiological conditions on the same species.

By investigating the spectra of cell envelopes and cytoplasmic extracts separately, distinguishable spectroscopic features of the microbial spectra can be identified. Tentative assignment of absorbance bands to their corresponding functional groups and the possible biomolecules is presented in Table I, based on experimental results achieved in this study and is in good agreements with the literature.11, 13–15

Table I. Tentative Assignment of Absorbance Bands in the IR Spectra of Microbial Cellsa
Wavenumber (cm−1)Functional Group AssignmentPossible Biomolecule Contributors
  • a

    References are listed by superscript.

  • ν: Stretch; δ: deformation

295713ν(CH3) asymmetricFatty acids
291913ν(CH2) asymmetricFatty acids
287211ν(CH3) symmetricFatty acids
285211ν(CH2) symmetricFatty acids
1790–175015ν(C[DOUBLE BOND]O) affected by Cl, etc.Not clear
174111ν(C[DOUBLE BOND]O)Lipid esters
170811, 13ν(C[DOUBLE BOND]O), H-bondedRNA, DNA
∼169511Amide I band components resulting from antiparallel pleated sheets and β-turnsProteins
∼168511  
∼167511  
∼167015ν(C[DOUBLE BOND]N)RNA/DNA bases
∼165511Amide I of α-helical structureProteins
∼163711Amide I of β-sheetsProteins
154811Amide IIProteins
151511ShoulderProteins
145713δ(CH2)Lipids, proteins
141515C[BOND]O[BOND]H in-plane bendingCarbohydrates, DNA/RNA backbone, proteins
140214δC(CH3)2 symmetricLipids, carbohydrates, proteins
131214Amide IIIProteins
128414Amide IIIProteins
124014ν(P[DOUBLE BOND]O) asymmetricPhospholipids
∼116013, 14δ(COP), ν(CC), δ(COH)DNA and RNA backbones
∼112013, 14ν(CC) skeletal trans conformationDNA and RNA backbones
1200–100011C[BOND]O[BOND]C, C[BOND]O dominated by ring vibrationsCarbohydrates
108511, 14ν(P[DOUBLE BOND]O) symmetricDNA and RNA, phospholipids
107613, 14ν(CC) skeletal cis conformationDNA and RNA backbones
900–80015C[DOUBLE BOND]C, C[DOUBLE BOND]N, C[BOND]H in ring structureNucleotides

Primary C[DOUBLE BOND]O stretching band of esters is at ∼1741 cm−1, which was observed in the IR spectrum (Figure 3). However, the hydrogen bond can have profound effects on the C[DOUBLE BOND]O stretching mode and can alter the position of this band by 30 cm−1.15 Therefore, the band at 1708 cm−1 (Figure 3) probably represents the C[DOUBLE BOND]O stretching mode, which might be due to DNA/RNA.11

thumbnail image

Figure 3. Spectroscopic features of E. coli O157:H7. (a) Cell envelopes, (b) cytoplasmic extract, and (c) and whole cell.

Download figure to PowerPoint

Another small peak could observed in the range between 1790 and 1750 cm−1; where Cl or other atoms with strong electronegativity is attached to C[DOUBLE BOND]O, the IR band would be presented in this region. However, it is questionable whether acidic chlorides were present in a microbial cell. The physical meaning of this peak is still not clear.

A comparison of the spectra of cytoplasmic extracts, cell envelopes, and whole cells of E. coli O157:H7 is shown in Figure 3. General patterns were similar for the other three bacteria (S. enteritidis, S. typhimurium, and E. coli O26) analyzed. When the spectra were measured for the whole cells, the signals were mostly from the cell walls—hence, the whole cell spectra was very similar to the cell envelope spectra. Only a small amount of the contributions from other cellular components present in the background inevitably interfered with the cell envelope signals and resulted in peak widening. It is believed that more details of the spectroscopic features might be revealed by investigating the spectra of cytoplasmic extract and cell envelopes separately compared to the spectra of the whole organisms. Figure 3 clearly shows that this belief is well based: the difference in the spectra of the cytoplasmic extract and cell envelopes and the compositional basis of these differences can be identified. The most distinguishable features of the spectra relate to the amide I and II bands, which are dominant in the spectrum of cell envelope and are less prominent in the cytoplasmic extract. Because the types of biomolecules present in the cytoplasm are much larger in numbers than that in the cell envelope, the bands observed for the cytoplasmic extract are broader, due to superposition of neighboring peaks, and it was extremely difficult to relate the individual contributions to specific compounds.

The band at ∼1400 cm−1, presumably due to superposition of C[BOND]O[BOND]H in-plane bending (1415 cm−1) and C(CH3)2 stretching (1402 cm−1), was the strongest in the cytoplasmic extract (Figure 3b). Carbohydrates, DNA/RNA backbone, and proteins all contribute to these bands, given the fact that amide I and II bands are relatively weaker here; this band may represent significant contributions from carbohydrates and DNA/RNA, which are abundant in the cytoplasm. It should be noticed that some authors6, 11 assigned this band to the C[DOUBLE BOND]O stretching in COO—it would thus be hard to explain why this peak was so prominent in the cytoplasmic extract.

Another noticeable difference was that at ∼1160 cm−1, the δ(COP), δ(CC), and δ(COH) groups, which represent the contribution due to DNA and RNA backbone, was clearly identifiable in the spectrum of the cytoplasmic extract, but less conspicuous in the cell envelope. Also below 1000 cm−1, in the region dominated by nucleic acid signals, the absorbance in the cell envelopes was almost zero. On the other hand, significant absorbance could be observed in this region in the cytoplasm (Figure 3a and b).

The relative absorbance intensity due to DNA, RNA, and lipids constituents (normalized against amide I) observed at 1670 cm−1 was much larger in the cytoplasmic extract than the cell envelope. Obviously this is due to the significant contribution of DNA and RNA. Another distinct feature is that in the spectrum of cell envelope, the asymmetric P[DOUBLE BOND]O stretch at 1250 cm−1, representing the contribution of phospholipids from cell membrane, was more prominent than the δ(COP) band at 1160 cm−1 while in the spectrum of cytoplasmic extract, the opposite was observed due to the contribution from DNA and RNA backbone (Figure 3a and b).

In the wavenumber range of 900–800 cm−1, absorbance was mostly due to the C[DOUBLE BOND]C and N[DOUBLE BOND]C groups in the nucleotide. It can be seen that almost no peaks can be identified in the spectrum of the cell envelopes (Figure 3a). In the spectrum of the cytoplasmic extract in Figure 3b, a very broad band was observed due to the superposition of contributions from the various DNA and RNA segments and contribution from some carbohydrates. Also, a sharp peak at 1100 cm−1 in the spectrum of the cytoplasmic extract was observed due to the characteristic absorbance of C[BOND]O[BOND]C ring.

Discriminant analysis of the four bacteria based on the spectra of cell envelope and cytoplasm is given in Figure 4. As expected, the spectroscopic features of the cell envelopes gave better discriminant power. This observation is consistent with the fact that all antibody binding sites are located at the surface of the cell envelopes, and serve as major markers for the identification of microbial species, and as the sole indicator of serogroups or serotypes for the same species.

thumbnail image

Figure 4. Differentiation of 4 bacteria using the FTIR spectra of (a) cytoplasmic extract and (b) cell envelope.

Download figure to PowerPoint

FTIR Characterization of Bacterial Growth

The growth of the bacterium in general can be described by an initial lag phase followed by an exponential phase or log phase; at the time when the nutrients in the medium can no longer support the exponential growth of the bacterium, an equilibrium is reached constituting the stationary phase. After stationary phase, when the nutrients are depleted the bacterium starts to die out, depicting the decline phase. During growth, the condition of the bacterium changes continuously and parallels the continuous physiological and compositional changes experienced by the cells during growth. The discussion to follow will attempt to characterize the growth from a spectroscopic perspective.

Shigella boydii was investigated in this study and its changes were monitored by FTIR microspectrometry to establish a relation between the changes in the cell characteristics during growth. Figure 5 shows the growth curve of S. boydii for a period of up to 48 h. Discriminant analysis was performed using WIN-DAS (Wiley & Sons, Chichester, UK. 1998 version) to identify the spectroscopic features that are dependent on the growing time, or the physiological conditions during the different growth phases.

thumbnail image

Figure 5. Growth curve using optical density for S. boydii.

Download figure to PowerPoint

Figure 6 shows the spectra of the bacterium, whole cells, and cytoplasmic extract, measured at different time periods during growth. It can be seen that the spectral features of the whole cells (predominantly cell envelopes) did not show significant changes through the time period (from 6 up to 48 h), other than the observed relative absorbance intensity of fatty acids (2700–3100 cm−1) normalized against the amide I band, which decreased continuously. On the other hand, since the composition of the cell envelope experiences fewer changes compared to the cytoplasmic extract as time elapsed, changes in the spectra of the cytoplasmic extract were prominent during growth. The spectra of the cytoplasmic extracts show dramatic changes at the band corresponding to the phosphate backbone of nucleic acids (∼1160 cm−1), a continuous increase until a maximum was reached at 36 h, depicting the rapid growth mode of the bacterium during the growth phase up to 36 h. This is further supported by the discriminant analysis (Figure 7), where the clusters formed from the spectra of the cytoplasmic extract were in good agreement with the growth curve. Three distinguishable clusters of supergroups can be easily identified: cluster 1 (6 h) denotes that the bacterium was still in lag phase; cluster 2, from 6 to 24 h, that he bacterium was in log phase,; and cluster 3, from 36 to 48 h, that the bacterium was in stationary phase. Spectral data measured for the whole cells did not produce distinct results, demonstrating that the spectra of the cytoplasmic extract could be used to study the development of cells and to distinguish cells of the same species under different physiological conditions.

thumbnail image

Figure 6. Spectra of S. boydii whole cells (a) and cytoplasmic materials (b) at growing time of 6, 9, 12, 24, 36, and 48 h.

Download figure to PowerPoint

thumbnail image

Figure 7. Chemometric analysis of the spectra of S. boydii (a) whole cells and (b) cytoplasmic extract at growing time of 6, 9, 12, 24, 36, and 48 h.

Download figure to PowerPoint

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSION
  7. REFERENCES

It was demonstrated in this study that the spectroscopic fingerprints of microbial cells obtained by the microspectroscopic approach provide important structural and compositional information of microbial cells of Salmonella (enteritidis and typhimurium), E. coli (serotype O26, O27, and O157:H7), Yersinia enterocolitis, and Shigella boydii; and can be used to differentiate these microorganisms down to the strain level. Cytoplasmic extract and cell envelopes of E. coli O157:H7 as well as other microorganisms displayed distinguishable features in their respective FTIR spectra that can be attributed to their compositional and structural differences, i.e., high DNA/RNA content in cytoplasm, and used as indicators to characterize the microorganism and to monitor the compositional and structural changes during cell growth. Spectral features of the cell envelopes of the microorganisms due to the specificity of surface proteins, i.e., antibody binding site, had unique biochemical signatures and served as a good discriminator of the antigens considered. On the other hand, the spectral features of the cytoplasmic extract reflect the physiological, compositional, and metabolic changes experienced by the microbes during growth. FTIR spectroscopy could thus serve as a powerful nondestructive tool to monitor cell growth and to extract physiological, compositional, and structural information of microorganisms during their life cycle.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS AND DISCUSSION
  6. CONCLUSION
  7. REFERENCES
  • 1
    Naumann, D.; Helm, D.; Labischinski, H. Nature 1991, 351, 8182.
  • 2
    Curk, M. C.; Peladan, F.; Hubert, J. C. FEMS Microbiol Lett 1994, 123, 241248.
  • 3
    Kümmerle, M.; Scherer, S.; Seiler, H. Appl Environ Microbiol 1998, 64, 22072214.
  • 4
    Timmins, E. M.; Howell, S. A.; Alsberg, B. K.; Noble, W. C.; Goodacre R. J Clin Microbiol 1998, 36, 367374.
  • 5
    Schmitt, J.; Udelhoven, T.; Naumann, D.; Flemming, H.-C. In Infrared Spectroscopy: New Tool in Medicine, Proceedings of SPIE; Mantsch, H. H., Jackson, M., Eds.; Bellingham, WA, 1998; Vol 3257, pp 236244.
  • 6
    Maquelin, K.; Kirschner, C.; Choo-Smith, L.-P.; van den Braak, N.; Endtz, H. Ph.; Naumann, D.; Puppels, G. J. J Microbiol Methods 2002, 51, 255271.
  • 7
    Yu, C.; Irudayaraj, J.; Debroy, C.; Schmilovtich, Z.; Mizrach, A. J Food Sci, in press.
  • 8
    Kneipp, K.; Kneipp, H.; Kartha, V. B.; Manoharan, R.; Deinum, G.; Itzkan, I.; Dasari, R. R.; Feld, M. S. Phys Rev E: Stat Phys, Plasmas, Fluids, Relat Interdiscip Top 1998, 57, R628.
  • 9
    Lasch, P.; Haensch, W.; Lewis, E. N.; Kidder, L. H.; Naumann, D. Appl Spectrosc 2001, 56(1), 19.
  • 10
    Lasch, P.; Naumann, D. Cell Mol Biol 1998, 44(1), 189202.
  • 11
    Naumann, D. In Encyclopedia of Analytical Chemistry; Meyers, R. A., Ed.; John Wiley & Sons Ltd.: Chichester, UK, 2000, pp 102131.
  • 12
    Kadner, R. J. In Escherichia coli and Salmonella; Neidhardt, F. C., Ed.; ASM Press: Washington, DC, 1996, pp 5887.
  • 13
    Mordechai, S.; Salman, A.; Argov, S.; Cohen, B.; Erukhimovitch, V.; Goldstein, J.; Chaims O.; Hammody, Z. In Biomedical Spectroscopy: Vibrational Spectroscopy and Other Novel Techniques, Proceedings of SPIE; Mahadevan-Jansen, A., Puppels, G. J., Eds.; SPIE, 2000, Vol 3918, pp 6677.
  • 14
    Lucassen, G. W.; Caspers, P. J.; Puppels, G. J. In Infrared Spectroscopy: New Tool in Medicine, Proc of SPIE; Mantsch, H. H., Jackson, M., Eds.; Bellingham, WA, SPIE, 1998, Vol 3257, pp 5261.
  • 15
    Socrates, G. Infrared and Raman Characteristic Group Frequencies, Tables and Charts, 3rd ed.; John Wiley & Sons: Chichester—New York—Brisbane—Toronto, 2001, p 123.
  • 16
    Lasch, P.; Petras, T.; Ullrich, O.; Backmann, J.; Naumann, D.; Grune, T. J Biol Chem 2001, 27(12), 94929502.
  • 17
    O'Haver, T. C. In Modern Fluorescence Spectroscopy; Wehry, E. L., Ed.; Plenum Press: New York, 1976.