2-DE study on natural resistance to intracellular parasites
Electrophoresis 1998, I Y , 1325-1331
Hana Kovaiova' Danuta Radzioch* Marian Hajd6ch3 Milada SirovA4 Viclav Blaha' AleS Macela' JiH Stulik' Lenka Hernychovzi'
1325
Natural resistance to intracellular parasites: A study by two-dimensional gel electrophoresis coupled with multivariate analysis
Natural resistance to Mycobacterium bovis bacillus Calmette-GuCrin (BCG) is determined by the Bcg gene (Nrampl), which is exclusively expressed by mature macrophages. The Nrampl gene is a dominant autosomal gene that 'PurkynG Military Medical Academy, has two allelic forms; r confers resistance and s confers susceptibility to infection with intracellular pathogen. Although the wide range of pleiotropic immuHradec KrhlovC, Czech Republic nological effects of the Nrampl gene has been described, the exact mechanism 'McGill Centre for the Study of of its action remains elusive. In this study we searched for differentially Host Resistance, Montreal, Canada expressed proteins that might provide clues in the studies on Nrampl gene 3Faculty of Medicine, Palacky University, Olomouc, Czech Republic function. We performed two-dimensional gel electrophoresis of cellular proteins prepared from a BlOR macrophage line derived from mice carrying the 4Academy of Science of the Czech r allele of the Nrampl gene, BlOS macrophages carrying the s allele, and BlORRepublic, Prague, Czech Republic Rb macrophages transfected with Nrampl-ribozyme. The classification of protein patterns and selection of distinct proteins characteristic of r or s allelecarrying macrophages was performed using the principal component analysis. We found differential expression of four proteins with the following isoelectric pointlmolecular weight (pIlM,) in BlOR macrophages compared to BlOS and BlOR-Rb macrophages: 6.6125, 7.0/22,9.1131.5, and 5.318.5. The protein 7.0122 has been identified as Mn-superoxide dismutase and the best candidate for protein ~6.6125seems to be Bcl-2 according to the immunoblot analysis. When the splenic macrophages carring the r or s allele were analyzed, the changes in relative abundance for proteins 6.6125 and p7.0122 were satisfactorily reproduced. Overall, the two identified proteins are important in the regulation of intracellular redox balance and the regulation of apoptosis in macrophages, respectively. Our findings may suggest their possible biological role in the innate immunity against intracellular pathogens.
1 Introduction Protective immunity against infection with intracellular bacterial pathogens combines the mechanisms of innate resistance and acquired immune response [l]. In the mouse, the Bcg/Zty/Lsh locus has been implicated in innate resistance to several microbial pathogens including Mycobacterium bovis BCG [2] and other taxonomically and antigenically unrelated microorganisms such as Salmonella typhimurium [3] and Leishmania donovani [4]. Positional cloning and subsequent analysis of Nrampl identified this gene as a serious candidate for the Bcg/ Ity/Lsh locus [5,6]. Nrampl is expressed by macrophages in two allelic forms, resistant (r) and susceptible (s), with a Gly-169 to Asp-169 substitution that is associated with susceptibility to infection [7]. The establishment of Nrampl knockout and transgenic mice provided definitive evidence that the natural resistance to infection with intracellular parasites is controlled by the Nrampl gene [8, 91. Studies on the functional and phenotypic activaCorrespondence: Dr. Hana Koviiovi, Institute for Immunology, Purkyne Military Medical Academy, Tfebelska str. 1575, 500 01 Hradec KralovB, Czech Republic (Tel: +420-49-521-0833; Fax: +420-49-5513018; E-mail:
[email protected]) Abbreviations: Mn-SOD, manganese superoxide dismutase; Nramp, natural resistance-associated macrophage protein; PCA, principal component analysis, ROI, reactive oxygen intermediates Keywords: Two-dimensional polyacrylamide gel electrophoresis / Natural resistance / Genetic control / Redox balance / Ion transport
0 WILEY-VCH
Verlag GmbH, 69451 Weinheim, 1998
tion of macrophages derived from Bcg' and Bcg' mice demonstrated a wide range of pleiotropic effects of this host resistance locus [lo, 1I]. Macrophages from innately resistant mice exhibited an increased state of activation in comparison to their susceptible counterparts. Unfortunately, a specific role for Nrampl in the control of these pleiotropic effects has not been defined yet and the exact function of Nrampl remains elusive.
In the present study, we assessed the association of r allele of the Nrampl gene with differential protein expression in .macrophages. We took advantage of the availability of macrophage cell lines derived from bone marrow of congenic mice carrying either the r or s allele of Nrampl, and a macrophage cell line in which Nrampl expression was abrogated by antisense Nrampl RNAribozyme transfection. High resolution two-dimensional electrophoresis (2-DE) was performed to separate macrophage proteins according to their charge and size. The global protein patterns were analyzed by computerassisted image analysis using Melanie I1 software (version 2.1, Bio-Rad). Principal component analysis (PCA) was applied in order to objectively classify relatedness of the cell specimens in regard to their protein composition. This analysis allowed us to select proteins differentially expressed in macrophages carrying the r allele of the Nrampl gene compared to their susceptible controls carrying either the s allele of the Nrampl gene or the Nrampl gene abrogated by tranfection with Nrampl-ribozyme. 0173-0835/98/0809-1325$17.50+.50/0
1326
H.
KOVA~OVA
et
Electrophoresis 1998, 19, 1325-133 1
UL
2 Materials and methods 2.1 Materials
cell pellet was lysed as described above. The total protein concentration in the samples was measured using Bio-Rad protein concentration assay kit (Cat. No. 5000006) and slightly modified according to Ramagli and Rodriguez [14].
Immobiline dry strip (IPG) pH 3-10 NL, 18 cm, carrier ampolytes Ampholines, pH 9-1 1, and bromophenol blue were from Pharmacia Biotech (Uppsala, Sweden), parrafin highly liquid, urea and glycerol were from Merck 2.3 2-DE (Darmstadt, Germany); Resolyte carrier ampholytes pH The protein separations were performed as described by 4-8 were purchased from BDH (Poole, UK); 3-([3-chola- Hochstrasser et al. [15] and Stulik et al. [16]. Samples midopropyl]dimethylammonio)-2-hydroxyl-l-propanesul- equal to 100 1.18 of proteins were loaded on the IPG fonate (CHAPS), Tris base, Tris-HC1, agarose, and iodo- strips by adding into the sample cups positioned on acetamide were from Sigma (St. Louis, MO, USA); di- cathodic edges of the strips. The isoelectric focusing thiothreitol (DTT), acrylamide, piperazinediacrylamide with immobilized sigmoid pH 3.5-10 gradient was car(PDA), ammonium persulfate (APS), and N,N,NN,-tetra- ried out in a Multiphor apparatus with a 5000 V power methylethylenediamine (TEMED) were from Bio-Rad supply. The second dimension was done in 9-16% poly(Hercules, CA); sodium dodecyl sulfate (SDS) and gly- acrylamide gradient gels using Protean I1 xi 2D multi cine were from Fluka (Buchs, Switzerland); interferon cell. Protein spots were visualized by sensitive ammogamma was from Genzyme Scandic (Prague, Czech niacal silver staining [17]. Republic); fetal calf serum, LPS free, was purchased from BioCom (Brno, Czech Republic). 2.4 Image processing 2.2 Cell lines and sample preparation
Macrophage cell lines BlOR (Nrampl r allele) and BlOS (Nrampl s allele) were derived by immortalizing bone marrow macrophages from BIO.A. Bcg‘ and BIO.A (Bcg‘) mice congenic in Bcg locus [12]. These macrophage lines faithfully reflect the morphological and functional properties of native macrophages from BCG-resistant and -susceptible mouse strains. Macrophage cell line BlORRb was prepared by stable transfection of the BlOR line with the construct encoding Nrampl-antisense RNAcontaining ribozyme sequences under the control of retroviral promoter [ 131. Cells were grown in Dulbecco’s modified essential medium containing 4.5 g/L glucose supplemented with 2 mM L-glutamine, 100 U/mL penicillin, 100 mg/mL streptomycin and 10% heat-inactivated fetal calf serum. For 2-DE, cell lines were cultured in 1 mL aliquots at a density of 1 X lo6 cells/mL in cell culture medium at 37°C and 5 % CO, for 18 h. Where indicated, 20 U/mL of mouse recombinant interferon gamma was added to the culture to prepare interferon-stimulated samples: BlORi, BlOSi, BlOR-Rbi. Following the 18 h incubation with or without interferon, the cells were washed three times with ice-cold phosphate buffered saline (PBS), the supernatants were carefully removed and cell pellets were lysed with 60 yL of lysis buffer containing urea (8 M), CHAPS (4% w/v), Tris (40 mM), DTT (65 mM), 2% v/v Ampholines pH 9-11 and a trace of bromophenol blue. The samples were centrifuged for 5 min at 12 000 rpm. Adherent splenic cells from B1O.A. Bcg‘ and B1O.A mice were prepared as follows: single cell suspensions from spleens were prepared and suspended in RPMI 1640 culture medium supplemented as above and containing 1% of fetal calf serum. The cell suspensions at a density of 2 X lo7 cells/mL were incubated on Petri dishes at 37°C in 5 % CO, for 3 h. Nonadherent cells were then removed by repeated washing in PBS. Adherent cells were collected using a cell scraper and cell debris was removed by washing in ice-cold PBS. The viability of cells was checked by trypan blue exclusion tests. The
Silver-stained gels were scanned using a laser densitometer (Personal Densitometer, Molecular Dynamics, 4000 x 5000 pixels, 12 bitdpixel; stored on 16 bits) generating 20-megabyte images. The images were then transferred to the SUN workstation for analysis with Melanie 11, version 2.1 software (Bio-Rad). For each gel the spots were detected and quantified automatically using default spot detection parameters from Melanie 11. Manual spot editing was performed in agreement with the visual inspection of the gels. Quantitation of spots was done in terms of their relative optical density (O/oOD), i.e., digitized staining intensity of an individual spot divided by the sum of staining intensities of all spots in the image and multiplied by 100. The relative optical density was chosen to correct for differences in protein loading and gel staining. The gels were matched in terms of their spot positions with BlOR gel as the reference gel.
2.5 Data analysis To evaluate quantitative changes in protein expression in macrophage samples, the analysis of 2-DE gel data was performed using PCA [18]. This multifactorial statistical approach makes it possible to reduce the number of observed abundant variables (protein spots) whilst preserving as much as possible of the original data. The principal components are measures of truly different “dimensions” in the data set with lack of abundant correlations. The 760 spots for this study were chosen on the basis of the criterion of their presence on all gels and their O/o OD served as the source for the initial matrix with four rows (gels) and 760 columns (spots). By choosing this matrix, it can be shown that principal components are calculated by finding eigenvalues and eigenvectors of correlation matrix [18]. These statistical calculations were performed using Statgraphic software version 4.2.
3 Results We performed 2-DE protein analysis of macrophage lines BlOR and BlOS carrying either the Nrampl r or s
Electrophoresis 1998, 19, 1325-1331
allele, respectively, as well as the BlOR-Rb macrophage line in which the Nrampl mRNA translation was disrupted by transfection of the antisense Nrampl RNAribozyme. An important advantage of this experimental set is the identical genetic background except for the 10 cM Bcg gene interval (the Nrumpl gene r or s allele). Since these macrophages represent a homogenous clonal population, sample-to-sample variability is minimal. Therefore, this experimental model is a powerful tool for both the detection and identification of proteins differentially expressed in macrophages carrying the r allele of the Nrampl gene compared to macrophages carrying the s allele of the Nrampl gene. About 1600 polypeptide spots were detected in each of the gels in the separating area of the 3.5-10 pH range and molecular mass 8-200 kDa range using Melanie I1 software. The polypeptide pattern of BlOR macrophages is shown in Fig. 1. Approx-
2-DE study on natural resistance to intracellular parasites
1327
imately 60% of the total number of polypeptide spots were successfuly matched in all analyzed macrophage samples. Reproducibility of 2-DE gels seems to be SUEcient as demonstrated by both closely matching results and confirmation of described protein alterations obtained from the second independently prepared set of gels. 3.1 PCA analysis
The analysis of such a large amount of data would be difficult by classical statistical means; therefore, we applied PCA to quantitative 2-DE gel data. To remove a major source of variance, which is usually the variation in spot abundance, we used relative instead of absolute OD values of the analyzed spots. The variances of the first three components are shown in Table 1. The first three
Figure I . Two-dimensional gel of silver-stained poLypeptides from B 10R macrophages. The proteins differentially expressed in BlOR macrophages compared to BlOS or BlOR-Rb macrophages are denoted by arrowheads. The pllM, index is shown for the proteins expressed at the higher level in BlOR macrophages.
1328
H.
K O V ~ ~ O V Pet
ai.
Electrophoresis 1998, 19, 1325-1331
Table 1. Variances of principal components Principal component 1 Principal component 2 Principal comoonent 3
Yo of variance
Cumulative percentage
40.48 35.61 23.91
40.48 76.09 100.00
components cover 1000/0of total variance of the data set. Figure 2 portrays the spatial distribution of gel samples using three-dimensional PCA-space projection. It is possible to distinguish two main groups of protein patterns. The group on the left side of the first principal component consists of BlOR macrophages carrying the r allele of Nrumpl while the group on the right side is represented by BlOS and BlOR-Rb macrophages. Relative nearness for each gel in PCA-space illustrates similarity. A larger distance between analyzed samples indicates dissimilarity in polypeptide abundance. PCA is not only suitable for grouping the gels according to similarity in their protein patterns but also for determination of typical polypeptide spots in the analyzed gels. The mathematical procedure allows simultaneous projection of gel samples and eigenvectors for most distinct proteins present in the same space defined by the first two principal components with largest eigenvalues (Fig. 3). The positions of gels with respect to the proteins with significant eigenvector values for the first principal component indicate which spots characterize the gels best. In Fig. 1 and 4, these "discriminant spots" are highlighted on BlOR and BlOS gels, respectively, and they are named according to their pI/M,: p6.6/25, p7.0/22, p9.1/31.5, and ~5.318.5.The relative abundance of these proteins in each macrophage sample is documented in Fig. 5. Two additional samples, BlOSi and BlOR-Rbi, are also included in this picture. We prepared these gels in order to increase reproducibility and to confirm protein changes that were selected by PCA. These samples were processed in the same way as the previous samples except for PCA calculating. However, the O/o OD of selected proteins was determined using Melanie I1 again and the values are presented in the same histogram. The relative abundance of p6.6125 and p7.0/22 proteins was higher in macrophages carrying the r allele of the Nrampl gene compared to macrophages carrying the s allele of the Nrampl gene. On the contrary, the relative abundance of p9.1131.5 and p5.3/8.5 proteins was lower in BlOR macrophages compared to BlOS and BIOR-Rb macrophages.
analysis after 2-DE separation while the candidacy of Bcl-2 is supported by quantitative alterations among macrophage samples on immunoblot after SDS-PAGE [19]. For other differentially expressed proteins in BlOR
4
3 -
Component 1
Figure 2. The spatial distribution of two-dimensional gel analyses of
BIOR, BlORi, BlOS and BIOR-Rb macrophages in three-dimensional PCA space. 1
"
'
1
"
'
1
"
'
1
"
'
1
"
~
I
7 - 0
6
-
*BlOS
h
a--i
ru cl C Ql
C
0
a
1 -
E
0 0
3.2 Manganese superoxide dismutase
The protein p 7.0/22 expressed more efficiently in macrophages carrying the resistant allele has been identified as the manganese superoxide dismutase (Mn-SOD) by matching with the reference protein database generated using U937 myelomonocytic cell line in SWISS2DPAGE (http://www.expasy.ch/). There was full compatibility of our gels with gels available in SWISS2DPAGE due to essentially the same experimental protocols of the Geneva laboratory and ours. The best possible candidate for ~6.6125seems to be Bcl-2 (or a protein closely homologous to it) whose theoretical value for pI/M, is 6.65/26 by calculating in Swiss-Prot. The identity of Mn-SOD is confirmed by immunoblot
-1
-
I
-4.4
-2.4
-8.4
1.e
Component
1.6
5.5
1
Figure 3. Biplot for the first two principal components showing simultaneous projection of gels and eigenvector values for most distinct polypeptide spots.
Electrophoresis 1998, 19, 1325-1331
2-DE study on natural resistance to intracellular parasites
1329
Figure 4. Two-dimensional gel of silver-stained polypeptides from BlOS macrophages. The proteins differentially expressed in BlOR macrophages compared to BlOS or BlOR-Rb macrophages are denoted by arrowheads. The pl/M, index is shown for the proteins expressed at the higher level in BlOS macrophages.
compared to BlOS macrophages we have found no matching candidate in the SWISS-2DPAGE database. 3.3 Fresh adherent spleen cell As a final step of our investigation we attempted to reproduce these findings using fresh adherent splenic cells carrying the r or s allele. We performed separation as well as image analysis in the same manner as described above and detected changes in relative abundance for ~6.6125 and p7.0122 proteins although the changes were less pronounced (37 and 20%, respectively) than in the case of macrophage lines (Fig. 6 ) .We assume that the reason for the less distinct differences may be the heterogeneity of adherent splenic cells compared to macrophage lines. No diference in the abundance of the p5.318.5 and p9.1131.5 proteins was detected in the adherent splenic cells of BIO.A. Bcg' and BIO.A mice.
4 Discussion 4.1 Principal components
Using 2-DE analysis of cellular proteins we were able to measure simultaneously the abundance of a large number of gene products. This large number of protein spots, i.e., variables, rendered both the evaluation of quantitative alterations and interpretation difficult. To solve this problem we utilized the multifactorial method known as PCA. It allows one to ignore the abundant measurements and to focus on the relevant differences in protein patterns analyzed. Instead of multidimensional space we can restrict ourselves to the first two and three components and represent results graphically by biplot and PCA space, respectively. The original variables are usually represented in biplots by vectors which indicate the proportion of the variability explained by two prin-
1330
Electrophoresis 1998, 19, 1325-1331
H. Kovaiova et al.
BBlOR OBlORi BlOS mBlOSi BlOR-Rb BlOR-Rbi
09 0.8
07
0,6
8
0.5
s
04
03
0.2 0.1
0
~6.6125
p7.0/22
p5 318.5
~9.1131.5
proteins
Figure 5. The adundance of the four selected proteins in BlOR, BlOS and BlOR-Rb macrophages and in macrophages stimulated by interferon gamma (BIORi, BIOSi, BlOR-Rbi). The values of relative optical density were obtained using Melanie I1 software. The average values calculated from two independently prepared gels are shown.
.A
I
n ""
08
07 0.6
0
05
s
04
0
tally induced infections when small doses of attenuated organisms are used [21]. It is of interest to realize that certain virulence factors produced by bacteria, for example sulfatide and lipoarabinomannan, can inhibit phagosome-lysosome fusion [22], deactivate formation of reactive oxygen intermediates (ROI), scavange ROI [23, 241, and alter cell signaling [25]. Indirectly, it points out the relevance of these steps in host cell response immediately after primary interaction with a pathogen. The r allele of the Nrampl gene increases, independently of T-cells, the ability of macrophages to control growth of intracellular organisms. It has been suggested that macrophages carrying the r allele are at a heightened state of activation, demonstrated by many pleiotropic functions acquired more efficiently by these macrophages compared to susceptible counterparts [13,26-281. Our attention has been focused on the analysis of changes in macrophage proteins in relation with expression of either the r or s allele of the Nrampl gene. A previous attempt to search for the product of the Nrampl gene according to the criterion that it should be present in macrophages carrying the r allele of this gene but not in macrophages carrying the s allele [29] or gene knockout failed. A possible explanation is the expected low abundance of Nrampl protein that remains hidden among undetectable proteins while working with whole cell lysates. We assume that separation of the membrane fraction and improvements in 2-DE of membrane proteins will reveal the Nrampl protein.
03
4.3 Nrampl coregulated proteins
02
01
0 pE 6125
p5.3/8.5
p7 OR2
pe.1131.5
proteins
Figure 6. The adundance of the four selected proteins obtained from
congenic B1O.A.Bcg' and BIO.A mice. The values of relative optical density were obtained using Melanie I1 software.
cipal components. The direction of the vectors shows their relative "weights" for the principal components [ 181. Vector values are especially important in the proposal and resolution of the possible model or mechanisms related to similarities or differences of studied samples This approach proved to be beneficial. Macrophages carrying the r allele of the Nrampl gene have been satisfactorily distinguished from the macrophages carrying the s allele or Nrampl ribozyme transfected counterparts. 4.2 Host genetics
Host genetics provides a major regulatory interface that influences the outcome of infection. The identification of the Nrampl gene as a candidate gene for the Bcg locus [S, 61 has been formally proven by construction of Nrumpl-knockout mice [8]. In murine models, it has been shown that the cell type responsible for phenotypic expression of the Bcg gene is the mature macrophage [20]. The resistance depends on the route and dose of antigen. The resistance to infection with M . bovis BCG in Bcg' mice is most pronounced early during experimen-
At present, we have identified a group of Nrampl coregulated proteins. One of the differentially expressed proteins might be Bcl-2, the second identified protein is Mn-SOD. The recent results of Supek et al. [30], showing that the Nrampl yeast homolog SMFI is the manganese transporter, match with our observations demonstrating alterations of Mn-SOD in macrophages carrying the r versus s allele of Nrampl. Mn-SOD is an enzymatic antioxidant that catalyzes the dismutation of superoxide radicals. It is located in the mitochondrial matrix near the electron transport chain [31]. Bcl-2, a prototype of a growing family of proteins that inhibit apoptosis, is associated with the outer mitochondrial membrane and endoplasmic reticulum. Although the mechanism of action of Bcl-2 is not known, there is some evidence to suggest that Bcl-2 may inhibit apoptosis by regulating ROI and calcium fluxes in the cell [32-341. It is still controversial whether or not the oxidative stress is a critical part of the apoptotic process. Oxidative stress might induce the program of cell death. While participating in the apoptotic process, it may also affect enzymes such as cysteine proteases and endonucleases that are dependent on reduced protein thiols for their activity 135, 361. It is also known that ROI, as well as thiol modulation, directly alter Mn-SOD expression [37, 381. ROI are also involved in the induction of Mn-SOD by interferon gamma, tumor necrosis factor alpha, and interleukin 1 [39]. Taking into consideration these facts, it is intriguing to speculate about possible functions of the Nrampl protein. On one hand, it might be involved in modulation of intracellular redox balance by influencing membrane potential and/or membrane electron transport systems
Elecfrophoresis 1998, 19, 1325-1331
that are sources of ROI. On the other hand, it may contribute to the induction of complementary antioxidant protective systems such as Mn-SOD. The phagocytes expressing the r allele of the Nrampl gene may utilize controlled oxidative stress for alterations of redoxsensitive signal transducing pathways such as deactivation of tyrosine phosphatases in interferon gamma signaling [40], activation of NF-XB [41], and activation of hck tyrosine kinase [42]. These phagocytes could also cope with oxidative stress, which, in turn, might be reflected in attenuation of their apoptotic pathway. The attenuation in the apoptosis of infected phagocytes appears to be one of the strategies of the immune system to preserve antigen-presenting cells [43, 441. Our suggestions about possible functions of the Nrampl protein are in agreement with the results of Barton et al. [45] who showed increases in respiratory burst and nitric oxide pathways after in vitro transfection of the resistant form of Nrampl cDNA into cell line RAW264.7. It also does not exclude the possibility of Vidal et al. [5] and Nathan [46] that the Nrampl protein functions as a nitrite transporter. The control of growth of ingested bacteria and its killing would take place within macrophage phagolysosomes where toxic ROI and reactive nitrogen intermediates (RNI) are present. ROI alone may be insufficient to kill ingested bacteria [47], but ROI combined with RNI can significantly enhance killing of bacteria [48,49]. We thank J. MichaliEkova, A . Firychova, J. Zakova, M. Safa?ova, and A . Janos’t’akova for excellent technical assistance. We are indebted to Proj I. Lefkovits at Basel Institute for Immunology for critically reading the manuscript. We are grateful for constant encouragement by Proj D. Hochstrasser at the University Hospital Geneva, as well as Proj E. Skamene at the McGill Centre for the Study of Host Resistance in Montreal. This work was supported by funds from the Ministry of Defense (H. Kovafova, grant “Gene”),Ministry of Health, Czech Republic (M. Hajduch, grant 3449-3), and MRC grant MT10707 ( 0 . Radzioch). Received March 27, 1997
5 References Fearon, D. T., Locksley, R. M., Science 1996, 272, 50-54. Gros, P., Skamene, E., Forget, A,, J. Immunol. 1981, 127, 2417-2421. Plant, J., Glynn, A. A., Clin. Exp. Immunol. 1979, 37, 1-6. Bradley, D. J., Clin. Exp. Immunol. 1977, 30, 130-140. Vidal, S. M., Malo, D., Vogan, K., Skamene, E., Gros, P., Cell 1993, 73, 469-485. Barton, C. H., White, J. K., Roach, T. I. A,, Blackwell, J. M., J. Exp. Med. 1994, 179, 1683-1687. Malo, D., Vogan, K., Vidal, S., Hu, J., Cellier, M., Schurr, E., Fuks, A., Bumstead, N., Morgan, K., Gros, P., Genomics 1994,23,51-61. Vidal, V. S., Tremblay, M., Govoni, G., Gauthier, S., Sebastiani, G., Malo, D., Skamene, E., Olivier, M., Joty, S., Gros, P., J. Exp. Med. 1995, 182, 655-666. Govoni, G., Vidal, S., Gauthier, S., Skamene, E., Malo, D., Gros, P., Infect. Immun. 1996, 64, 2923-2929. Skamene, E., Immunobiol. 1994, 191, 451-460. Blackwell, J. M., Barton, C. H., White, J. K., Roach, T. I. A., Shaw, M.-A,, Whitehead, S. H., Mock, B. A., Searle, S., Wiliams, H., Baker, A.-M., Immunol. Lett. 1994, 43, 99-107.
2-DE study on natural resistance to intracellular parasites
1331
[12] Radzioch, D., Hudson, T., Boule, M., Barrera, R., Urbance, J. W., Varesio, L., Skamene, E., J. Leukocyte Biol. 1991, 50, 263-272. [13] Radzioch, D., Kramnik, I., Skamene, E., Circulatory Shock 1995, 44, 115-120. [14] Ramagli, L. S., Rodriguez, L. V., Electrophoresis 1985, 6, 559-563. [15] Hochstrasser, D. F., Frutiger, S., Paquet, N., Bairoch, A,, Ravier, F., Pasquali, C., Sanchez, J.-C., Tissot, J.-D., Bjellquist, B., Vargas, R., Appel, R. D., Hughes, G. J., Electrophoresis 1992, 13, 992-1001. [16] Stulik, J., Bures, J., Jandik, P., Langr, F., Kovarova, H., Macela, A,, Electrophoresis 1997, 18, 625-628. [17] Bjellqvist, B., Pasquali, C., Ravier, F., Sanchez, J.-C., Hochstrasser, D. F., Electrophoresis 1993, 14, 1357-1365. [18] Manly, B. F. J., Multivariate Statistical Methods: A Primer, Chapman and Hall, London 1986, pp. 59-71. [19] Kovaiova, H., Hajduch, M., Macela, A,, Electrophoresis 1997, 18, 2654-2660. [20] Gros, P., Skamene, E., Forget, A., J. Immunol. 1983, 127, 2417-242 1. [21] Forget, A., Skamene, E., Gros, P., Miailhe, A.-C., Turcotte R., Infect. Immun. 1981, 32, 42-47. [22] Clemens, D. L., Horwitz, M. A., J. Exp. Med. 1995, 181, 257-270. [23] Chan, J., Fan, X. D., Hunter, S. W., Brennan, P. J., Bloom, B. R., Infect. Immun. 1991, 59, 1755-1761. [24] Pabst, M., Gross, J., Brozna, J., Goren, M., J. Immunol. 1988, 140, 634-640. [25] Reinier, N. E., Immunol. Today 1994, 15, 374-381. [26] Denis, M., Forget, A., Pelletier, M., Skamene, E., J. Immunol. 1988, 140, 2395-2400. [27] Blackwell, J. M., Roach, T. I. A,, Atkinson, S. E., Ajioka, J. W., Barton, C. H., Shaw, M.-A., Immunol. Letters 1991, 30, 241-248. [28] Buschman, E., Skamene, E., Immunol. Letters 1988, 19, 199-210. [29] Vidal, S. M., Pinner, E., Lepage, P., Gauthier, S., Gros, P., J. Immunol. 1996, 157, 3559-3568. [30] Supek, F., Supekova, L., Nelson, H., Nelson, M., Proc. Natl. Acad. Sci USA 1996, 93, 5105-5110. [31] Bannister, J. V., Bannister, W. H., Rotilio, G., Crit. Rev. Biochem. 1987, 22, 111-121. [32] Hockenberry, D. M., Oltvai, Z. N., Yin, X.-M., Milliman, C. L., Korsmeyer, S. J., Cell 1993, 75, 214-251. [33] Lam, M., Dubyak, G., Chen, L., Nunez, G., Miesfeld, R. L., Proc. Natl. Acad. Sci. USA 1994, 91, 6569-6573. [34] Wolvetang, E. J., L a m , J. A., Moutsoulas, P., Lawen, A., Cell Growth Different. 1996, 7, 1315-1325. [35] Buttke, T. M., Sandstrom, P. A,, Immunol. Today 1994, 15, 7-10. [36] Orrenius, S., Nobel, C. S. I., Van den Dobbelsteen, D. J., Burkitt, M. J., Slater, A. F. G., Biochem. SOC.Transact. 1996,24, 1932-1038. [37] Warner, B. B., Stuart, L., Gebb, S., Wispe, J. R., Am. J. Physiol. 1996, 271, L150-L158. [38] Kumuda, C. D., Lewis-Molock, Y.,White, C. W., Mol. Cell. Biochem. 1995, 148, 45-75. [39] Harris, C. A,, Derbin, K. S., Hunte-McDonough, B., Krauss, M. R., Chen, K. T., Smith, D. M., Epstein, L. B., J. Immunol. 1991, 147, 149-154. [40] Lamb, P., Haslam, J., Kessler, L., Seidel, M. H., Stein, R. B., Rosen, J., J. Interferon Res. 1994, 14, 365-373. [41] Kretz-Remy, C., Mehlen, P., Mirault, M.-E., Arrigo, A.-P., J. Cell Biol. 1996, 133, 1083-1093. [42] Brumell, J. H., Burkhardt, A. L., Bolen, J. B., Grinstein, S . , J. Biol. Chem. 1996, 271, 1455-1461. [43] Durrbaum-Landmann, I., Gercken, J., Flad, H. D., Ernst, M., Infect. Immun. 1996, 64, 5384-5389. [44] Heidenreich, S., Otte, B., Lang, D., Schmidt, M., J. Leukocyte Biol. 1996, 60, 737-743. [45] Barton, C. H., Whitehead, S. H., Blackwell, J. M., Mol. Med. 1995, I , 261-279. [46] Nathan, C., Cell 1995, 82, 873-876. [47] O’Brien, S., Jackett, P. S., Douglas, B. L., Andrew, P. W., Microbial Pathogenesis 1991, 10, 199-207. [48] Beckman, J. S., Crow, P., Biochem. Sac. Trans. 1993, 21, 330-334. [49] Zhu, L., Gunn, C., Beckman, J., Arch. Biochem. Biophys. 1992,298, 452-457.
