Characterization of Chromobacterium violaceum pigment through a hyperspectral imaging system
© Gallardo et al.; licensee Springer. 2014
Received: 5 November 2013
Accepted: 30 December 2013
Published: 13 January 2014
In this paper, a comprehensive spatio-spectral and temporal analysis for Chromobacterium violaceum colonies is reported. A hyperspectral imaging (HSI) system is used to recover the spectral signatures of pigment production in a non-homogeneous media with high spectral resolution and high sensitivity in vivo, without destructing the sample. This non-contact sensing technique opens avenues to study the temporal growing of a specific section in the bacterial colony. Further, from a 580 [nm] and 764 [nm] spatio-spectral time series, a wild-type and mutant Chromobacterium violaceum strains are characterized. Such study provides quantitative information about kinetic parameters of pigment production and bacterial growing.
KeywordsChromobacterium Hyperspectral systems Real time sampling Pigment production
For decades, natural pigments have been extensively used in various fields of everyday life such as food production, textile industries, paper production, agricultural practices and researches, water science and technology (Arad and Yaron1992; Sirimanne et al.2006). Natural pigments not only have the capacity to increase the marketability of products, they also display advantageous biological activities as antioxidants and anticancer agents (Stahl and Sies2005; Kong et al.2010). Several intensely colored compounds have been isolated from certain microorganisms and from various environmental sources (Ahmad et al.2012). An example is the violacein pigment, a purple-colored dye produced by one of the strains of Chromobacterium violaceum, which is an indole derivative (Duran and Menck2001) with antitumoral, antibacterial, antiulcerogenic, antileishmanial, and antiviral activities. (Leon et al.2001; Duran et al.2007).
The detection of bacterial components like antibiotics, enzymes and secondary metabolites like pigments, is important in biotechnological industries (Steele and Stowers1991). Traditional methods are effective but can take hours or days to realize them. Besides, they are destructive as well as time consuming. On the other hand, molecular detection based techniques for bacteria identification are rapid, specific and sensitive. However, most are technically complicated and costly, and require well-trained specialists (Steele and Stowers1991; Lazcka et al.2007; Velusamy et al.2010). Now, molecular spectroscopy provides less invasive or even non-destructive methods that have been proven well-suited to investigate the chemical composition of biological systems (Cen and He2007).
Hyperspectral imaging (HSI) is an emerging technique that assimilates spectroscopy and imaging to provide both spectral and spatial information of a biological sample (Plaza et al.2009). Further, the potential that HSIs have for in vivo optical diagnostics have being exploited due to the non-invasive feature and the massive spatio-spectral information collected by the system (Vo-Dinh2004; Pisani et al.2013). In this sense, HSI systems record, for each spatial location being imaged, a set of hundreds of high spectral resolution images that jointly conform the spectrum of the biological sample. Such collections of images is known as hypercube (Borengasser et al.2008), from which the pertinent qualitative and quantitative information is obtained (Sankaran et al.2010; Siripatrawan et al.2011). Different optical techniques can be applied to generate the hypercube from biological samples, some of them are based on tunable optical filters (Gat2000), imaging spectrometers (Jun et al.2009), and coded hyperspectral imaging (Studer et al.2012), among other. In particular, push-broom hyperspectral cameras (PBHCs) are imaging devices that carried out the spectral decomposition by means of an optical process (Kim et al.2010). Further, PBHCs employ a scanning procedure to record spectral information of one spatial line at a time, with the ability to scan multiple batches of samples simultaneously by moving across the process line. Also, the spatial information is important for monitoring the sample as it can be used to extract its chemical mapping at different spatial points (Zavattini et al.2003; Singh et al.2010). In this regard, we can use HSI system to measure the spectral signature of chemical components, like bacterial pigments, inside the cells without altering the sample and its measurements even in a heterogeneous medium where spectrophotometric signal is not resolutive.
In this work, a spectral characterization of Chromobacterium violaceum pigment through a HSI is performed from the visible spectrum to part of the near infrared spectrum (400 nm to 1000 [nm] with 1,04 [nm] of spectral resolution). It is well known that a pigment changes the color of the reflected light as a result of selective color absorption. In this sense, the pigment is quantified associating its signal to the maximal optical absorption centered at 580 [nm] (Gerhardt,1970; Koch,1994). With HSI it is now possible to resolve signals and quantify the bacterial pigment with a remote, non-invasive and real-time procedure comparatively to a spectrophotometric measurement. Finally, a series of spectral signatures masking relevant information that can serve as an indicator of a specific bio-processes is obtained, i.e. pigment production and bacterial growing rate. In this case, despite using a small number of spectral bands, these are selected as a function of the amount of spectro-temporal information associated with the pigmentation process. For a different process, the evaluation of the spectral information can determine a larger number of spectral bands required for characterization. Hence, the amount of useful spectral information is directly related to the process being measured.
Materials and methods
The C. violaceum wild-type strain ATCC 31532 and C. violaceum CV026 (mini-Tn5 mutant of ATCC 31532) (McClean et al.1997) are cultured in Luria Bertani (LB) broth (10% peptone, 5% yeast extract, 5% NaCl, per 1 L distilled water) (Sambrook et al.1989). Solid bacterial medium is made by the addition of agar at a final concentration of 15 g/L.
The bacterial population and the pigment production is estimated by measuring absorbance of the culture using a spectro-photometer (UV-1700 UV-VIS Spectro-photometer) at different wavelengths (350-700 [nm]). Measurements are taken each one hour during lag phase and every 30 minutes in the exponential phase. For each measurement an aliquot of liquid culture is taken and the absorbance is determined every 50 [nm] (between 350 and 700 [nm]).
Spot plating and incubation
The HSI system employed in this work is the Photonfocus Hurricane 40 V10E PBHC. This camera is based on a CMOS FPA of 1024 × 1024 photo-detectors. Further, the input radiance is decomposed and converted into several spatio-spectral images through an optical process that allows us to obtain spectral bands between 400-1000 [nm], with a spectral resolution of 1.04 [nm]. Also, due to the nature of the acquisition process of PBHC, a mobile platform has been constructed to simulate and synchronize the scanning procedure.
Furthermore, to compensate for any noise source degrading the hyperspectral data, it is required to perform calibration measurements. To this end, we have employed a quartz tungsten halogen (QTH) calibration lamp, which guarantees an uniform and continuous spectral illumination between 200-2500 [nm]. Also, a Spectralon SRT- 99-120 is used as a diffuse reflectance target that ensures a reflectance of 99% between 250-2000 [nm].
As shown in Figure2, the diffuse target lies on a petri dish holder to increase the reflectivity of the semi-transparent agar and colonies. Further, during the scanning process, each line of the target scene acquired receives the same angle of illumination, then we can ensure that all spatial points were illuminated under the same conditions. The sample measurement scheme is displayed in Figure1, where the input radiance is generated by the light reflected from the agar. Finally, the dimensions of the hypercubes produced after scanning the whole target scene are 1024 spatial pixels, 574 spectral bands, and 1000 temporal lines.
Spectral reflectance recovery
where the suffixes i, λ j , and k denote, respectively, the spatial location at the scene, the spectral band, and the temporal sample, where i = 1,…,P, j = 1,…,Q′, and k = 1,…,S. The variable R(i,λ j ,k) represents the input reflectance collected at the λ j th spectral band by the ij th photo-detector at the k th sample time. The terms a(i,j) and b(i,j) represent, respectively, all the multiplicative and additive noise sources corrupting the output of a PBHC. The additive term V(i,j,k) is known as the temporal white noise associated with the readout electronic for the ij th photo-detector. However, in most of the cameras the white noise V(i,j,k) is negligible compared to the term b(i,j), so it can be disregarded. Finally, the parameter r(λ j ) represents the system spectral response, i.e. the slit, the diffraction grid response and the spectral responsivity of the detector material.
Kinetic parameter of pigment production and bacterial growth
Hyperspectral imaging measurementa
580 [nm] (pigment production + bacterial growth)
5.28 ± 1.10
0.137 ± 0.029
29.77 ± 6.85
0.025 ± 0.006
764 [nm] (bacterial growth)
7.57 ± 1.14
0.094 ± 0.014
26.93 ± 7.31
0.028 ± 0.008
580 [nm]/764 [nm] (pigment production)
OD measurement b
2.93 ± 0.65
0.249 ± 0.055
8.4 ± 1.23
0.08 ± 0.011
The grown parameters g and k for both strains at 764 [nm] are calculated. At this wavelength only information of bacterial growing is obtained. Although the wild-type strain grows faster than the mutant strain both have comparable velocities k. This is in agreement with the values found in literature and those obtained with the spectrophotometric measurements (see Table1).
The violacein production can be quantified using a spectrophotometer, after an extraction and purification process (Rettori and Duran1998; Lu et al.2009). This procedure fails to identify pigment remotely and not allows to follow the process in time. To follow the pigment production without destroying the bacterial cells we use a Hyperspectral Imaging System. This system provide spectral and spatial information of the Chromobacterium violaceum enabling us to obtain kinetic parameters of bacterial growth and pigment production.
The violacein production can be quantified at 580 [nm] by subtracting the contribution of the bacterial cell components at 764 [nm] which resulted in k pigment = 0.043. Therefore, the results are in good agreement with the values calculated using optical density (OD) at 600 [nm] (see Table1) and with those found in the literature (Corpe1964).
In this work, a real time, non invasive and simple process to quantify bacterial pigment production trough HSI system is reported. The PBHC camera produces a hypercube with spatial and spectral information in function of time without destroying the sample. Hence we can analyze multiple samples with high wavelength resolution in a single experiment. Based on the process, the technique can be extensible to multiple colored samples at the same time. This optical technique could be useful for the analysis of complex samples such as bacterial biofilms, environmental samples or non-culture bacteria (Cortizo and de Mele2000). This can be applied to quantify another biological pigments obtained from natural source without alter the environmental conditions. For example, studies can be done to quantify photosynthetic pigments, like chlorophylls or carotenoids, presents in some plants or bacteria. Also this technique can be used in the pigment determination in fruits or other exportation products. The advantage of this technique is the decomposition of the components of the biological system in a set of wavelengths each one associated to a specific spatio-temporal character. This allows to get new insights of a particular process in a complex media. Also, since we have detected the specific wavelengths that allows to characterize the bacterial pigment production and growing process, the next step is developing a filter-based imaging system with the aim to increases the temporal resolution of the measurements. That is, designing a monochromatic camera that includes a filter tuned at the wavelength of interest, discarding the scanning stage and acquiring the spatio-spectral image that exhibits greater variation in time (Nishino et al.2013). Further, such devices do not require a scanning stage, which could increase the temporal resolution of our measurements.
This work was supported from PIA-CONICYT PFB0824. M.J. Gallardo acknowledge support from FONDECYT 3140167. J.P. Staforelli acknowledge support from FONDECYT 11110145. Pablo Meza is supported by a CONICYT PhD Scholarship. Ignacio Bordeu is supported by a CONICYT MSc Scholarship.
- Ahmad WA, Yusof N, Nordin N, Zakaria ZA, Rezali MF: Production and characterization of violacein by locally isolated Chromobacterium violaceum grown in agricultural wastes. Appl Biochem Biotechnol 2012, 167: 1220–1234. doi:10.1007/s12010–012–9553–7 10.1007/s12010-012-9553-7View ArticlePubMedGoogle Scholar
- Arad SM, Yaron A: Natural pigments from red microalgae for use in foods and cosmetics. Trends in Food Sci Technol 1992, 3(0):92–97. doi:10.1016/0924–2244(92)90145-MView ArticleGoogle Scholar
- Borengasser M, Hungate WS, Watkins RL: Hyperspectral remote sensing: principles and applications. CRC Press, Series Remote Sensing Applications Series 2008, 130. doi:10.1201/9781420012606.ch3Google Scholar
- Cen H, He Y: Theory and application of near infrared reflectance spectroscopy in determination of food quality. Trends in Food Sci Technol 2007, 18(2):72–83. doi:10.1016/j.tifs.2006.09.003 10.1016/j.tifs.2006.09.003View ArticleGoogle Scholar
- Corpe WA: Factors influencing growth and polysaccharide formation by strains of Chromobacterium Violaceum . J Bacteriol 1964, 88(5):1433–1441.PubMed CentralPubMedGoogle Scholar
- Cortizo M, de Mele M: Preliminary characterization of thin biofilms by optical microscopy. BIOFOULING 2000, 15(4):253–260. doi = 10.1080/08927010009386316 10.1080/08927010009386316View ArticleGoogle Scholar
- Duran N, Menck C: Chromobacterium violaceum : A review of pharmacological and industiral perspectives. Critic Rev Microbiol 2001, 27(3):201–222. doi:10.1080/20014091096747 10.1080/20014091096747View ArticleGoogle Scholar
- Duran N, Justo GZ, Ferreira CV, Melo PS, Cordi L, Martins D: Violacein: properties and biological activities. Biotechnol Appl Biochem 2007, 48(3):127–133. DOI:10.1042/BA20070115 10.1042/BA20070115View ArticlePubMedGoogle Scholar
- Gat N: Imaging spectroscopy using tunable filters: a review. 2000, 4056: 50–64.Google Scholar
- Gerhardt P: Methods for general and molecular bacteriology. Washington, DC: American Society for Microbiology; 1994:518–518.Google Scholar
- Jun W, Kim M, Lee K, Millner P, Chao K: Assessment of bacterial biofilm on stainless steel by hyperspectral fluorescence imaging. Sensing Instrum Food Qual Saf 2009, 3(1):41–48. DOI:10.1007/s11694–009–9069–1 10.1007/s11694-009-9069-1View ArticleGoogle Scholar
- Kim TH, Kong HJ, Kim TH, Shin JS: Design and fabrication of a 900–1700 nm hyper-spectral imaging spectrometer. Optics Commun 2010, 283(3):355–361. DOI:10.1016/j.optcom.2009.10.022 10.1016/j.optcom.2009.10.022View ArticleGoogle Scholar
- Koch AL: Turbidity measurements of bacterial cultures in some available commercial instruments. Anal Biochemis 1970, 38(1):252–259. DOI:10.1016/0003–2697(70)90174–0 10.1016/0003-2697(70)90174-0View ArticleGoogle Scholar
- Kong KW, Khoo HE, Prasad KN, Ismail A, Tan CP, Rajab NF: Revealing the power of the natural red pigment lycopene. Molecules 2010, 15(2):959–987. DOI:10.3390/molecules15020959 10.3390/molecules15020959View ArticlePubMedGoogle Scholar
- Lazcka O, Campo FJD, Munoz FX: Pathogen detection: A perspective of traditional methods and biosensors. Biosensors and Bioelectronics 2007, 22(7):1205–1217. DOI:10.1016/j.bios.2006.06.036 10.1016/j.bios.2006.06.036View ArticlePubMedGoogle Scholar
- Leon LL, Miranda CC, Souza AOD, Durán N: Antileishmanial activity of the violacein extracted from Chromobacterium violaceum . J Antimicro Chemotherapy 2001, 48(3):449–450. DOI:10.1093/jac/48.3.449 10.1093/jac/48.3.449View ArticleGoogle Scholar
- Lu Y, Wang L, Xue Y, Zhang C, Xing XH, Lou K, Zhang Z, Li Y, Zhang G, Bi J, Su Z: Production of violet pigment by a newly isolated psychrotrophic bacterium from a glacier in Xinjiang, China. Biochem Eng J 2009, 43(2):135–141. 10.1016/j.bej.2008.09.009View ArticleGoogle Scholar
- McClean KH, Winson MK, Fish L, Taylor A, Chhabra SR, Camara M, Daykin M, Lamb JH, Swift S, Bycroft BW, Stewart GSAB, Williamsl P: Quorum sensing and Chromobacterium violaceum : exploitation of violacein production and inhibition for the detection of n-acylhomoserine lactones. Microbiol 1997, 143(12):3703–3711. DOI:10.1099/00221287–143–12–3703 10.1099/00221287-143-12-3703View ArticleGoogle Scholar
- Nishino K, Nakamura K, Tsuta M, Yoshimura M, Sugiyama J, Nakauchi S: Optimization of excitation–emission band-pass filter for visualization of viable bacteria distribution on the surface of pork meat. Opt Express 2013, 21(10):12,579–12,591. DOI:10.1364/OE.21.012579.http://www.opticsexpress.org/abstract.cfm?URI=oe-21–10–12579 10.1364/OE.21.012579View ArticleGoogle Scholar
- Pisani M, Zucco M, Caricato V, Egidi A: Hyperspectral imaging: a tool for biological measurements. In 16th International Congress of Metrology Edited by: Filtz J-R, Larquier B, Claudel P, et Favreau J. -O. 2013, 14007–14007. doi =10.1051/metrology/201314007,http://dx.doi.org/10.1051/metrology/201314007View ArticleGoogle Scholar
- Plaza A, Benediktsson JA, Boardman JW, Brazile J, Bruzzone L, Camps-Valls G, Chanussot J, Fauvel M, Gamba P, Gualtieri A, Marconcini M, Tilton JC, Trianni G: Recent advances in techniques for hyperspectral image processing. Remote Sensing Environment 113 2009, Supplement 1(0):S110-S122. DOI:10.1016/j.rse.2007.07.028View ArticleGoogle Scholar
- Rettori D, Duran N: Production, extraction and purification of violacein: an antibiotic pigment produced by Chromobacterium violaceum . World J Microbiol Biotechnol 1998, 14(5):685–688. 10.1023/A:1008809504504View ArticleGoogle Scholar
- Sambrook J, Fritsch E, Maniatis T: Molecular Cloning: a laboratory manual. Cold Spring Harbor Laboratory Pr, N.Y., Cold Spring Harbor Laboratory; 1989.Google Scholar
- Sankaran S, Ehsani R, Etxeberria E: Mid-infrared spectroscopy for detection of huanglongbing (greening) in citrus leaves. Talanta 2010, 83(2):574–581. DOI:10.1016/j.talanta.2010.10.008 10.1016/j.talanta.2010.10.008View ArticlePubMedGoogle Scholar
- Singh CB, Jayas DS, Paliwal J, White ND: Identification of insect-damaged wheat kernels using short-wave near-infrared hyperspectral and digital colour imaging. Comput Electron Agric 2010, 73(2):118–125. DOI:10.1016/j.compag.2010.06.001 10.1016/j.compag.2010.06.001View ArticleGoogle Scholar
- Sirimanne P, Senevirathna M, Premalal E, Pitigala P, Sivakumar V, Tennakone K: Utilization of natural pigment extracted from pomegranate fruits as sensitizer in solid-state solar cells. J Photochemis Photobiol A: Chemis 2006, 177(2–3):324–327. DOI:10.1016/j.jphotochem.2005.07.003 10.1016/j.jphotochem.2005.07.003View ArticleGoogle Scholar
- Siripatrawan U, Makino Y, Kawagoe Y, Oshita S: Rapid detection of Escherichia coli contamination in packaged fresh spinach using hyperspectral imaging. Talanta 2011, 85(1):276–281. DOI:10.1016/j.talanta.2011.03.061 10.1016/j.talanta.2011.03.061View ArticlePubMedGoogle Scholar
- Stahl W, Sies H: Bioactivity and protective effects of natural carotenoids. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease 2005, 1740(2):101–107. DOI:10.1016/j.bbadis.2004.12.006 10.1016/j.bbadis.2004.12.006View ArticleGoogle Scholar
- Steele DB, Stowers MD: Techniques for selection of industrially important microorganisms. Ann Rev Microbiol 1991, 45(1):89–106. DOI:10.1146/annurev.mi.45.100191.000513 10.1146/annurev.mi.45.100191.000513View ArticleGoogle Scholar
- Studer V, Bobin J, Chahid M, Mousavi HS, Candes E, Dahan M: Compressive fluorescence microscopy for biological and hyperspectral imaging. Proc Nat Acad Sci 2012, 109(26):E1679-E1687. DOI:10.1073/pnas.1119511109 10.1073/pnas.1119511109PubMed CentralView ArticlePubMedGoogle Scholar
- Velusamy V, Arshak K, Korostynska O, Oliwa K, Adley C: An overview of foodborne pathogen detection: In the perspective of biosensors. Biotechnol Adv 2010, 28(2):232–254. DOI:10.1016/j.biotechadv.2009.12.004 10.1016/j.biotechadv.2009.12.004View ArticlePubMedGoogle Scholar
- Vo-Dinh T: A hyperspectral imaging system for in vivo optical diagnostics. Eng Med Biol Mag, IEEE 2004, 23(5):40–49. DOI:10.1109/MEMB.2004.1360407 10.1109/MEMB.2004.1360407View ArticleGoogle Scholar
- Zavattini G, Vecchi S, Leahy R, Smith D, Cherry S: A hyperspectral fluorescence imaging system for biological applications. In Nuclear Science Symposium Conference Record. 2003 IEEE; 2003:942–946, Vol.2. DOI:10.1109/NSSMIC.2003.1351850Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.