- Email: [email protected]
Gene expression in activated brain microglia: identification of a proteinase inhibitor that increases microglial cell number
Molecular Brain Research 56 Ž1998. 99–107
Research report
Gene expression in activated brain microglia: identification of a proteinase inhibitor that increases microglial cell number Smita Thakker-Varia a
a,)
, Stella Elkabes a , Charles Schick b, Gary A. Silverman b, Lang Peng a , Ann C. Sherwood a , Ira B. Black a
Department of Neuroscience and Cell Biology, UniÕersity of Medicine and Dentistry of New Jersey-Robert Wood Johnson Medical School, Piscataway, NJ 08854, USA b Joint Program in Neonatology, Department of Pediatrics, HarÕard Medical School, Children’s Hospital, Boston, MA 02115-5737, USA Accepted 27 January 1998
Abstract Microglia, the intrinsic immune cells of the central nervous system, are activated in a variety of inflammatory brain diseases in which they play a pathogenetic role. However, mechanisms underlying activation are largely unknown. To begin elucidating molecular mechanisms associated with activation, we characterized the pattern of gene expression in virtually pure dissociated microglial cultures, using RT-PCR differential display. Microglia were activated with bacterial lipopolysaccharide ŽLPS., a traditional stimulant, and the profile of gene expression was compared to that in basal, control cultures. Activation resulted in altered expression of six genes. The cDNAs were isolated, sequenced and characterized. Homology searches identified three novel genes, and two that exhibited very high sequence similarity to the gene encoding squamous cell carcinoma antigen ŽSCCA.. SCCA Ž1 and 2. are tandemly arranged genes that encode two serine proteinase inhibitors Žserpins.. SCCA has been detected exclusively in cancer cells, and is a plasma marker for squamous cell carcinoma. Immunoblot analysis indicated that gene expression was accompanied by a 5-fold increase in the synthesis of SCCA protein in LPS-activated microglia. To assess potential biological actions of the SCCA serpins, SCCA1 protein was added to cultures. SCCA1 altered microglial morphology, and elicited a dramatic, 5-fold increase in cell number within 72 h. The effects appeared to be cell-specific, since the protein had no effect on other cell types: cortical astrocytes and neurons from cortex or basal forebrain were unaffected. We tentatively conclude that SCCA1 may play a cell-specific role in increasing cell number, a critical early step in microglial activation and brain inflammation. More generally, differential display of genes in the microglial model system may help define patterns of expression associated with CNS disease, thereby identifying pathogenetic mechanisms and new therapeutic targets. q 1998 Elsevier Science B.V. Keywords: Microglial activation; Lipopolysaccharides; Differential display; Squamous cell carcinoma antigen and serine proteinase inhibitor
1. Introduction In the central nervous system ŽCNS., response to injury and disease involves glial activation. After injury, microglial activation, in particular, takes place in a stereotypic pattern. Activation is characterized by proliferation and migration of microglia to the site of injury, accompanied by characteristic morphological, immunophenotypical and functional changes w6,9,21,26x. Activation may exert
) Corresponding author. UMDNJ-Robert Wood Johnson Medical School, 679 Hoes Lane, Piscataway, NJ 08854-5635. Fax: q1-732-2354990; E-mail: [email protected]
0169-328Xr98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 9 - 3 2 8 X Ž 9 8 . 0 0 0 3 4 - 5
both beneficial andror deleterious effects. Beneficial responses include tissue repair by secretion of growth factors, destruction of invading microorganisms and phagocytosis of potentially damaging debris. However, activated microglia also release potentially cytotoxic substances, such as free radical intermediates, NO, proteases and cytokines w8,10,28,33x. Despite these responses of microglia to CNS injury, little is known about the mechanisms underlying activation. To understand the molecular genetics of microglial activation, it is important to determine associated patterns of gene expression. Delineation of gene expression governing activation may help define regulation of the components of activation that mediate beneficial and deleterious actions of microglia.
100
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
To begin assessing the molecular basis of microglial activation, we employed a model of virtually pure dissociated cell cultures of microglia. Bacterial lipopolysaccharide ŽLPS., a component of the cell wall of gram negative bacteria, and a classic microglial stimulator, was used to activate the cultures. Activation of microglial cultures with LPS modifies cellular functions, including release of cytokines, proteases and reactive oxygen intermediates w8,10,28,33x. To examine the profile of gene expression that regulates activation, we employed RT-PCR differential display. This technique identifies differentially expressed genes by comparing steady-state levels of mRNAs in cells under various conditions, independent of RNA prevalence. The entire spectrum of both abundant and rare mRNAs is screened using this method. We report that during microglial activation, a number of known and unknown genes are differentially regulated. We identified squamous cell carcinoma antigen ŽSCCA. partial cDNA that was specifically induced in activated microglia and was associated with an increase in SCCA protein. The SCCA family consists of two nearly identical proteins, SCCA1 and SCCA2 w27x. SCCA1 exhibited cell-specific biologic effects. Increased microglial cell number with altered morphology was evident after treatment with purified protein. Thus, using this microglial culture model of activation in conjunction with the PCR-based method, we have begun to define genes and their functions associated with microglial activation. Identification of activation-associated genes may lead to discovery of new physiologic and pathogenic mechanisms involved in brain inflammation, and development of new therapeutic targets.
2. Materials and methods 2.1. Isolation of microglia Microglia were isolated from postnatal day 2 Sprague– Dawley rats ŽHilltop, Scottsdale, PA.. Cortices were dissected. Special care was taken to remove all meninges and blood vessels during dissection to minimize contamination by blood monocytes and macrophages. Mixed glial cultures were grown by the modification of the method of McCarthy and DeVellis w20x and shaken for 2 h at 180 rpm at 378C. Microglia were isolated by differential adhesion on plastic dishes Ž; 10 min, 378C., followed by trypsinization. Microglial cultures were grown in N2 medium containing 10% heat-inactivated fetal bovine serum. The microglial cultures obtained were ) 98% pure, 2 and 5 days after plating. The purity and identification of the cultures was assessed immunocytochemically using the CD11brc ŽOX-42. and ED1 markers w4,5,11,12,23x.
2.2. Isolation of astrocytes To establish astrocyte cultures, postnatal day 2 Sprague–Dawley rats ŽHilltop. were used. Mixed glial cells were grown as described above. Oligodendrocytes and microglia were removed by shaking at 250 rpm overnight. The adhering astrocytes were dissociated and plated in Eagle’s minimum essential medium containing 15% heat-inactivated fetal calf serum w32x. The astrocytes were identified and confirmed for purity by immunostaining for glial fibrillary acidic protein ŽGFAP.. Oligodendrocytes and O-2A progenitor cells were identified using antibodies to myelin basic protein ŽMBP. and A2B5, respectively. 2.3. Isolation of neurons Pregnant Sprague–Dawley rats ŽHilltop. were used. Embryos were recovered at embryonic day 17 ŽE17.. The dissections were performed as previously described w15x but with some modifications. Cells were plated at a density of 700 000 cellsr35 mm dishes coated with poly-D-lysine. The cultures were maintained in serum-free fully defined medium w32x. This medium restricts the growth of glia and promotes virtually pure neuronal cultures. 2.4. RNA isolation and differential mRNA display Total RNA from pure microglial cultures was extracted by the guanidine isothiocyanate method w2x. After removal of chromosomal DNA contamination in total RNAs, differential mRNA display analysis was carried out as described by Liang and Pardee w19x using the RNA image kit as instructed by the manufacturer ŽGene Hunter, MA.. Control reactions were performed in the absence of reverse transcriptase. Total RNA Ž2 m g. was reverse transcribed in a 20 m l reaction mixture using a one-base anchored primer HT11 A. The cDNAs were then amplified by PCR in the presence of Ž 35 S. dATP, and in controls, water was substituted for cDNA. Each set of the reaction mixtures included an arbitrary 13-mer Ž5X AAGCTTGATTGCC 3X , 5X AAGCTTCGACTGT 3X or 5X AAGCTTTGGTCAG 3X . from the RNAimage kit as 5X primer and HT11 A as 3X primer. The conditions for PCR were as follows: denaturation at 948C for 30 s, annealing at 408C for 2 min, and extension at 728C for 30 s; reactions were subjected to 40 cycles. Radiolabeled PCR products were analyzed by denaturing polyacrylamide gel electrophoresis. The gels were dried on Whatman paper without fixing. Putative differentially expressed bands were excised, resuspended in 100 m l of H 2 O for 10 min at RT, boiled for 15 min followed by reamplification in the absence of isotope in a 40-cycle PCR w18,19x. Reamplified cDNAs ranging from 100 to 300 bp were used for cloning and as templates for random priming.
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
101
2.5. TA cloning and sequence analysis
2.9. Cell counts
Reamplified cDNA fragments were cloned directly into the plasmid vector pCR II using the TA cloning kit ŽInvitrogen, CA.. Sequences of the cDNAs were acquired on an Applied Biosystems 373A automatic sequencer as per manufacturer’s specifications ŽApplied Biosystems, Foster City, CA.. Comparisons of the unknown nucleotide sequences to known sequences in the Genbank and EMBL data bases was performed using BLAST program ŽGenetics Computer, Madison, WI..
To assess the total number of cells per 35 mm dish, three randomly selected, non-overlapping 4.75 mm2 strips were counted Ž1.48% of the culture dish area..
2.6. Antibodies
3. Results
OX-42 and ED1 monoclonal antibodies were purchased from Serotec ŽOxford, UK.. The rabbit polyclonal antibody to GST-SCCA Žglutathione S-transferase. was characterized and successfully used for western blot analysis of pure proteins w24x, cell lines expressing SCCA1 and SCCA2 and tissue extracts Žpersonal communication.. 2.7. Western blot analysis Microglial cells plated in equal numbers were solubilized in lysis buffer ŽPBS–1% Triton X-100.. Cell lysates were incubated for 10 min at 48C. The protein content was determined with the Coomassie plus protein assay kit ŽPierce Chemical.. Samples containing equal amounts of protein concentration were denatured in Laemmli’s sample buffer Žcontaining b-mercaptoethanol. for 5 min and subjected to 6–12% gradient SDS-PAGE w17x. The proteins were transferred to polyvinylidenedifluoride ŽPVDF. membranes ŽMillipore. which were blocked for 1 h with a 4% solution of dry milk powder in 0.2% Tween 20–PBS. The PVDF membranes were incubated at 48C with rabbit polyclonal antisera against GST-SCCA Ž1:2000 diln. w24x for 1 h at RT or overnight at 48C. Membranes were washed, followed by 1 h incubation with HRP-conjugated IgG Ž1:500. at RT. The immunopositive bands were visualized by chemiluminescence using the ECL detection kit ŽAmersham, Arlington Heights, IL.. 2.8. Cell treatments Cells were grown in the absence or presence of the physiological dose of 100 ngrml LPS ŽSigma, St Louis, MI.. Cells were then harvested and used for RNA isolation or cell lysate preparation. Microglial cells, cortical astrocytes and neurons were incubated with 4.8 m grml of purified GST-SCCA1 Žglutathione S-transferase. protein for up to 5 days. The GST-SCCA1 fusion protein used was batch purified using glutathione–Sepharose 4B beads and was free of any bacterial contaminants. The cells were observed and monitored on daily basis. To confirm the specificity of SCCA1 effects, microglia were incubated with purified GST for 5 days, as controls.
2.10. Statistical analysis Data were analyzed by Student’s t test or one-way ANOVA.
3.1. Microglial culture model system To characterize the profile of gene expression associated with activation, we examined virtually pure microglial cultures. We used CD11brc ŽOX-42. and ED-1, specific markers for the microglialrmacrophage lineage, to define purity. Ninety-eight to ninety-nine percent of the cells expressed ED-1, and CD11brc ŽOX-42.. One to two percent of the cells were A2B5-positive, O-2A progenitor cells. No GFAP-positive cells were detected. The cultures were exposed to LPS Ž100 ngrml., for 24 h. The lowest LPS concentration that elicited changes in microglial trophic factor and receptor expression was chosen for the present studies Žunpublished results.. 3.2. Differential mRNA display To identify differentially expressed genes potentially involved in microglial activation, we compared patterns of mRNA expression in control cultures with those stimulated by LPS for 24 h. Three different combinations of primer sets were used to amplify subsets of mRNA ŽFig. 1.. The primers consisted of 1 anchored primer HT11 A, and three short arbitrary primers w5X AAGCTTGATTGCC 3X , Žlanes A and B., 5X AAGCTTCGACTGT 3X Žlanes C and D., or 5X AAGCTTTGGTCAG 3X Žlanes E and F.x. Amplified cDNA species in control and activated microglia were largely identical, indicating that the general gene expression profile remained unaltered. Nevertheless, specific differences were apparent with activation. Lanes A, C and E represent products from control cultures, and lanes B, D and F represent PCR products from LPSactivated cells ŽFig. 1.. Differential expression was observed in a number of cDNAs. We concentrated on qualitative differences; six representative bands were chosen for further analysis. These PCR products were prominent bands Žindicated by arrows in Fig. 1. and exhibited differential expression during LPS-activation. Activation was associated with both increased and decreased gene expression. The cDNAs from bands 2–6 exhibited differential upregulation of mRNA during LPSactivation: expression was detectable only in LPS-activated
102
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
samples and not in controls. The cDNA from band 1 reflected marked downregulation of mRNA during LPSactivation; mRNA expression was detectable in control samples, and was undetectable after activation. To validate the display results, the bands were recovered from the dried gel and reamplified using corresponding primer sets. The cDNA fragment from band 5 did not reamplify and was not pursued further. ŽFailure to reamplify this cDNA could have resulted from poor recovery of the cDNA from the gel or from ‘false positivity’.. cDNA products were cloned into the PCR II plasmid vector. The nucleotide sequence of these cDNAs was determined. At least five recombinant subclones for each selected band were systematically sequenced Ždata not shown.. All the cDNAs appeared to be AT rich, indicating that they corresponded to the 3X end of the mRNAs. Some of the cDNAs contained a putative polyadenylation signal w31x proximate to the upstream sequences complementary to the anchored oligo dT primers. Consequently, these cDNA fragments correspond to the 3X end of the mRNAs, as expected. 3.3. Sequence analysis and homology Homology searches revealed that two of the activated sequences corresponded to known genes, validating the display approach ŽTable 1.. The 196 and 200 bp fragments Žbands 2 and 3. corresponded to bases 1609–1707 Žwith 69% homology. of the published sequence of human squamous cell carcinoma antigen cDNA Žaccession no. S66896. w27x including the 3X untranslated sequence. SCCA is a member of the serine proteinase inhibitor superfamily. No known genes with significant homology to bands 1, 4, and 6 Ž176, 97 and 167 bp. were found in the Genbank or EMBL DNA data bases. 3.4. Expression of mature SCCA protein To begin assessing the biological significance of differential gene expression, we initially focused on SCCA. Western blot analysis was performed to determine whether the transcribed message for SCCA induced during activation was translated into mature protein. The expression of SCCA protein was markedly increased in LPS-activated microglial cell lysates, corroborating the results obtained
Table 1 Analysis of cDNA fragments identified by differential mRNA display Fig. 1. Differential display comparing RNAs from control and LPS-treated microglia cells. Total RNA was extracted from control Žlanes A, C, E. and LPS-treated microglial culture Žlanes B, D, F. and subjected to differential mRNA analysis. Autoradiograms of amplified PCR products are shown for three different primer combinations that identified distinct fragments Žarrows. with differential expression in the LPS-activated X group. Primer combinations included HT11 A as 3 primer for all reactions X X X and three arbitrary 13-mers Ž5 AAGCTTGATTGCC 3 , 5 AAGCTTCX X X X . GACTGT 3 or 5 AAGCTTTGGTCAG 3 as 5 primer.
Band number
Fragment size
Regulation
Sequence homology
1 2
176 bp 196 bp
down up
3
200 bp
up
4 6
97 bp 167 bp
up up
no homology human squamous cell carcinoma antigen human squamous cell carcinoma antigen no homology no homology
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
103
Fig. 2. ŽA. Immunoblot analysis of SCCA protein in control and LPS-treated microglia cells. Equal amounts Ž50 m g. of protein were loaded in each lane, electrophoresed and immunoblotted with rabbit polyclonal antisera specific for GST-SCCA. Potential SCCA–proteinase complex Ž; 72 kDa. and SCCA Ž; 42 kDa. bands Žindicated by arrows. show increased expression of both proteins in LPS-treated cells. Panel ŽB. shows the densitometric quantitation of the ; 42 kDa band. The results are expressed as mean "S.E.M. of three independent experiments Ž p - 0.006 by Student’s t test..
by differential display ŽFig. 2A.. Control samples were generated by incubating microglia in N2 medium containing 10% heat-inactivated fetal bovine serum, but no LPS. SCCA-specific protein of ; 42 kDa was detected in samples treated with LPS for 24 h Žlane 2.. The polyclonal antibody to the GST-SCCA conjugate also recognized a band migrating at ; 72 kDa w7,24x, corresponding in molecular mass to endogenous SCCA–proteinase complex. Serpins form SDS-stable complexes at a stoichiometry of 1:1, upon binding to their target proteinases w24x. Densitometric analyses ŽFig. 2B. of the samples from three different experiments are expressed as percent of control. A 5-fold increase of SCCA-specific protein in the LPS-treated cells was evident. Control cells exhibited barely detectable levels of SCCA Žlane 1.. Thus, both the mRNA for SCCA and its protein product were induced in microglia with activation by LPS.
was observed. This range of increases may be due to cell culture variation. No change was observed in control cultures treated with GST alone Žnot shown.. In addition to increased cell number, SCCA1-treated microglia exhibited marked morphological changes, consistent with activation. SCCA1 elicited a marked increase in microglial cell size. Cells assumed a flattened, circular
3.5. Biological effects of SCCA1 To assess the potential biological effects of SCCA family serpins Žconsisting of SCCA1 and SCCA2., on microglia, cells were treated with purified GST-SCCA1 or control GST protein. After 72 h of treatment with purified SCCA1, a dramatic, 5-fold increase in cell number was evident ŽFig. 3.. In subsequent experiments using different culture preparations, a 2- to 5-fold increase in cell number
Fig. 3. Effect of SCCA1 on microglial cell number. Sister cultures in triplicate were plated in 35 mm dishes, 4.8 m grml of purified GSTSCCA1 was added to the cultures after 3 h and the cell number was assessed 72 h after plating. A 5-fold higher cell number was seen in the SCCA1-treated cultures. Values are the mean of triplicate dishes "S.E.M. in a representative experiment. The experiment was repeated 3 times. )Significantly different from control Ž p- 0.0001. by Student’s t test.
104
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
105
appearance, or an amoeboid configuration, with diameters of 40–60 m m, characteristic of activated microglia ŽFig. 4A.. In contrast, GST control cells were small and spindle-shaped, with only rare circular forms. Nevertheless, a degree of heterogeneity was evident with SCCA1 stimulation. Large process-bearing cells and smaller clubshaped cells were also evident. Finally, SCCA1 exposure was associated with cytoplasmic granularity, principally associated with the large, circular cells. In contrast, GSTtreated cultures resembled untreated controls ŽFig. 4B..
ential display screening method involves PCR amplification of subsets of mRNAs and separation of the resulting cDNAs by gel electrophoresis. This technique is extremely sensitive and identifies both abundant and rare mRNAs. The combination of virtually pure microglial culture with differential display appears to constitute a fruitful approach to elucidating the molecular genetics of activation and inflammation.
3.6. Cell specificity of SCCA1 effects
Our initial screen identified 6 differentially regulated cDNAs after LPS stimulation, including both novel and previously defined genes. Out of six cDNAs, five were upregulated and one cDNA was downregulated. These cDNAs with altered expression may represent genes important for one or more of the functional components of activation. Further characterization of the novel genes identified in our study will clarify their potential roles. In this study we have focused on a known gene of interest, SCCA, and its relation to the activation process in pure microglial culture.
Since microglia synthesize and respond to SCCA, actions may represent autocrine andror paracine regulation. To begin ascertaining the specificity of SCCA1 for microglia, biological effects on other cell types were evaluated. We compared effects on astrocytes, basal forebrain neurons and cortical neurons. SCCA1 exhibited no effects on cell number or morphology in cultures of cortical astrocytes, or neurons from cortex or basal forebrain. After 48 h of treatment with SCCA1, no significant change in cortical astrocyte number was observed Žcontrol:1.36= 10 5 " 0.33; SCCA1 treated: 1.43 = 10 5 " 0.88.. Similarly, after 72 h of treatment with SCCA1, neurons from either basal forebrain Žcontrol: 1.0 = 10 5 " 2.03; SCCA1 treated: 1.2 = 10 5 " 0.88. or cortex Žcontrol: 5.4 = 10 5 " 2.92; SCCA1 treated: 6.2 = 10 5 " 0.67. exhibited no significant change in number. The morphology of the neurons was unaltered. Thus SCCA1 exerts cell-specific effects on microglia, not evident in other cell types.
4. Discussion 4.1. Microglial actiÕation alters the gene expression profile Microglial activation is a hallmark of diverse CNS disorders. Activated microglia are characterized by multiple morphological and functional changes, including chemotaxis, proliferation, phagocytosis and cytoskeletal rearrangements. The signaling mechanisms regulating these changes in microglia and the associated gene expression are largely undefined. To identify genes potentially involved in microglial activation, we used a potent stimulator, LPS, to activate virtually pure dissociated cultures of microglia. Differential mRNA display was used to delineate gene expression associated with activation. The differ-
4.2. OÕerall profile
4.3. Identification of SCCA Within 24 h of LPS treatment, a number of genes were upregulated. We identified a gene of the serine proteinase inhibitor family, squamous cell carcinoma antigen. The upregulation of this serpin Žsee below. was measurable within 24 h, a time course consistent with an inflammatory response. Furthermore, elevation of SCCA message was associated with a 5-fold increase in SCCA protein expression. This observation simultaneously served to validate the display approach, and indicate that increased SCCA gene expression is accompanied by elevation of the protein product. Activation of microglia by LPS, therefore, apparently increased the levels of SCCA by both transcriptional and translational mechanisms. Elevation of SCCA protein itself raised the possibility of biologically relevant effector activities. 4.4. SCCA protein increases microglial number To define potential functions for SCCA in the inflammatory response, specific biological effects of one of the two serpins, SCCA1 were examined. Purified SCCA1 protein induced a 5-fold increase in microglial cell number, suggesting possible autocrine or paracrine function. During inflammation and injury, microglial infiltration associated with increased cell number is observed at the
Fig. 4. Effect of SCCA1 on microglial morphology. Microglial cells were incubated with 4.8 m grml of purified GST-SCCA1 protein for up to 5 days and controls were incubated with buffer. Cells were monitored on a daily basis and counted after 72 h. In the presence of SCCA1, the cells exhibited altered morphology typical of activation with an increased cell number ŽB.. Arrows indicate circular amoeboid cells and arrowhead indicates large process-bearing cell. Typical smaller spindle-shaped morphology was seen in control cells ŽA. shown by undulating arrows. Scale bar equals 40 m m.
106
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
affected site. The increase may result from either chemotaxis, increased proliferation, andror enhanced survival. In our studies, increased cell number may have been attributable to increased mitosis, enhanced survival, or both w3x. We are presently examining these alternatives experimentally w3x. LPS-induced secretion of SCCA may enhance survival or enhance proliferation by inhibiting one or more serine proteinases inhibitory to microglia. 4.5. SCCA induces altered morphology Microglial cell morphology was significantly altered by purified SCCA1. SCCA1-treated cells were characterized by a large cell body with flattened, circular amoeboid-like morphology, consistent with activation. Conversion of resting microglia to an activated form occurs in response to injury or inflammation accompanied by characteristic changes in morphology. Proteases are essential to cytoskeletal rearrangement and cell adhesion properties associated with morphological restructuring. Increased microglial SCCA during inflammation may modulate these effects.
astrocytes, PN-1 is upregulated after transient ischemia w13x. PN-1 is also expressed in brain regions where continuous remodeling of neuronal connections takes place. Another serpin, a-1 antichymotrypsin is present in the Alzheimer amyloid plaques w30x and may facilitate deposition of b-amyloid peptide. By analogy with other serpins, SCCA1 may be involved in activating microglia by inducing proliferation. Increased expression of SCCA by activated microglia may also have a chemotactic function. The serpin-enzyme receptor complex found on surfaces of monocytes and neutrophils w14x may also be present on microglia, and SCCA might mediate neutrophil and or microglial chemotactic activity at injury sites. In conclusion, using differential display and an in vitro model of pure dissociated microglia, we have identified both known and novel genes that exhibit differential expression during activation. Among the known genes, SCCA, a member of the serpin family alters cell number.
Acknowledgements 4.6. SCCA1 effects are cell-specific The specificity of the effects of SCCA1 was examined using cortical astrocytes, as well as neurons from cortex and basal forebrain. In marked contrast to effects on microglia, SCCA1 did not alter cell number of the astrocytes or cortical or basal forebrain neurons. Moreover, this serine proteinase inhibitor did not alter cellular morphology of these cell types. Taken together, our findings of increased microglial number in the presence of SCCA1 and specific alteration in morphology, demonstrate that SCCA1 specifically and selectively affected microglia. SCCA is a member of the ovalbumin family of serine proteinase inhibitors. Recent studies have indicated that two nearly identical, tandemly arrayed SCCA genes are found in the genome w25x. Although the amino acid sequences of SCCA1 and SCCA2 are nearly identical, there are differences in reactive site loops suggesting differences in specificity for target proteinases. SCCA2 is effective against neutrophil cathepsin G and mast cell chymase w24x. Since inhibitory serpins regulate serine proteinases in diverse processes, including inflammation, cell migration and extracellular matrix remodeling w1,7,22,29x, SCCA may be an important regulator of increased proteolysis during microglial activation. SCCA serves as a serological marker for more advanced tumors of the cervix, lung and oropharynx w15,16x, but its functional role is not well defined either in normal or malignant cells. Although SCCA has not been previously associated with microglia, our results suggest that it may be an important regulator of microglial activation. A number of CNS protease inhibitors have been associated with disease processes, for example protease nexin-1 ŽPN-1. and a-1 antichymotrypsin. In reactive hippocampal
This work was supported by NISCD-35572-02 grant to S.E. and by NICHD-23315 and APA IBC-9501 grants to I.B.B. and by NIH CA 69331 to G.A.S.
References w1x R.W. Carrel, K.S. Aulak, M.C. Owen, The molecular pathology of the serpins, Mol. Biol. Med. 6 Ž1989. 35–42. w2x P. Chomczynski, N. Sacchi, Single step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extract, Anal. Biochem. 162 Ž1987. 156–169. w3x E. DiCicco-Bloom, W.J. Freidman, I.B. Black, NT-3 stimulates sympathetic neuroblast proliferation by promoting precursor cells, Neuron 11 Ž1993. 1101–1111. w4x C.D. Dijkstra, E.A. Dopp, P. Joling, G. Kral, The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulations in the rat recognized by monoclonal antibodies ED1, ED2 and ED3, Immunology 54 Ž1985. 589–599. w5x S. Elkabes, E.M. DiCicco-Bloom, I.B. Black, Brain microgliarmacrophages express neurotrophins that selectively regulate microglial proliferation and function, J. Neurosci. 16 Ž1996. 2508–2521. w6x J. Fulcrand, A. Privat, Neurological reaction secondary to Wallerian degeneration in the optic nerve of the postnatal rat: ultrastructural and quantitative study, Proc. Natl. Acad. Sci. U.S.A. 176 Ž1977. 189–224. w7x P.G.W. Gettins, P.A. Patston, S.T. Olson, Serpins: Structure Function and Biology, Molecular Biology Intelligence Unit, R.G. LandesrChapman & Hall, Austin, TX, 1996. w8x D. Giulian, Microglia, cytokines and cytotoxins: modulators of cellular responses after injury to the central nervous system, J. Immunol. Immunopharmacol. 10 Ž1990. 15–21. w9x D. Giulian, J. Chen, J.E. Ingeman, J.K. George, M. Noponen, The role of mononuclear phagocytes in wound healing after traumatic injury to adult mammalian brain, J. Neurosci. 9 Ž1989. 4416–4429. w10x D. Giulian, K. Vaca, M. Corpuz, Brain glia release factors with opposing actions upon neuronal survival, J. Neurosci. 13 Ž1993. 29–37.
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107 w11x M.B. Graeber, W.J. Streit, G.W. Kreutzberg, Axotomy of the rat facial nerve leads to increased CR3 complement receptor expression by activated microglial cells, J. Neurosci. Res. 21 Ž1988. 18–24. w12x M.B. Graeber, W.J. Streit, G.W. Kreutzberg, The third glial cell type, the microglia: cellular markers of activation in situ, Acta Histochem. 37 Ž1990. 157–160, Suppl. w13x M.C. Hoffmann, C. Nitsch, A.L. Scotti, E. Reinhard, D. Monard, The prolonged presence of glia-derived nexin, an endogenous protease inhibitor, in the hippocampus after ischemia-induced delayed neuronal death, Neuroscience 49 Ž1992. 397–408. w14x G. Joslin, G. Griffin, A. August, S. Adams, R. Fallon, R. Senior, D. Perlmutter, The serpin-enzyme complex ŽSEC. receptor mediates the neutrophil chemotactic effect of a-antitrypsin-elastase complexes amyloid-b peptide, J. Clin. Invest. 90 Ž1992. 1150–1154. w15x H. Kato, in: S. Sell ŽEd.., Serological Cancer Markers, Humana Press, Totowa, NJ, 1992, pp. 437–451. w16x H. Kato, T. Torigoe, Radioimmunoassay for tumor antigen of human cervical squamous cell carcinoma, Cancer 40 Ž1977. 1621–1628. w17x U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4 , Nature 227 Ž1970. 680–685. w18x P. Liang, L. Averboukh, A.B. Pardee, Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction, Nucleic Acids Res. 21 Ž1993. 3269–3275. w19x P. Liang, A. Pardee, Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction, Science 257 Ž1992. 967–971. w20x K.D. McCarthy, J. DeVellis, Preparation of separate astroglial and oligodendroglial cell cultures from rat cerebral tissue, J. Cell Biol. 85 Ž1980. 890–902. w21x Y. Murabe, Y. Ibata, Y. Sano, Morphological studies on neuroglia: III. Macrophage response and ‘microgliocytosis’ in kainic acid induced lesions, Cell Tissue Res. 218 Ž1981. 75–86. w22x J. Potempa, E. Korzus, J.T. Korzus, The serpin superfamily of proteinase inhibitors: structure, function, and regulation, J. Biol. Chem. 269 Ž1994. 15957–15960. w23x A.P. Robinson, T.M. White, D.W. Mason, Macrophage heterogeneity in the rat is delineated by two monoclonal antibodies MCR
w24x
w25x
w26x
w27x
w28x
w29x w30x
w31x w32x
w33x
107
Ox-41 and MRC Ox-42, the latter recognizing complement receptor type 3, Immunology 57 Ž1986. 239–247. C. Schick, Y. Kamach, A.J. Bartuski, S. Cataltepe, N.M. Schechter, P.A. Pemberton, G.A. Silverman Pemberton, Squamous cell carcinoma antigen 2 is a novel serpin that inhibits the chymotrypsin-like proteinases cathepsin G and mast cell chymase, J. Biol. Chem. 272 Ž1997. 1849–1855. S. Schneider, C. Schick, K. Fish, E. Miller, J. Pena, S. Treter, S. Hui, G. Silverman, A serine proteinase inhibitor locus at 18q21.3 contains a tandem duplication of the human squamous cell carcinoma antigen gene, Proc. Natl. Acad. Sci. U.S.A. 92 Ž1995. 3147– 3151. G. Stoll, B.D. Trapp, J.W. Griffin, Macrophage function during Wallerian degeneration in the rat optic nerve: clearance of degenerating myelin and Ia expression, J. Neurosci. 9 Ž1989. 2327–2335. Y. Suminami, F. Kishi, K. Sekiguchi, H. Kato, Squamous cell carcinoma antigen is a new member of the serine protease inhibitors, Biochem. Biophys. Res. Commun. 181 Ž1991. 51–58. C. Thery, B. Chamak, B. Mallat, Cytotoxic effect of brain macrophages on developing neurons, Eur. J. Neurosci. 3 Ž1991. 1155–1164. J. Travis, G.S. Salvesen, Human plasma proteinase inhibitors, Annu. Rev. Biochem. 52 Ž1983. 655–709. S. Wagner, J. Geddes, C. Cotman, A. Lau, D. Gurwitz, P. Isackson, D. Cunningham, Protease nexin-1, an antithrombin with neurite outgrowth activity, is reduced in Alzheimer disease, Proc. Natl. Acad. Sci. U.S.A. 86 Ž1989. 8284–8288. M. Wickens, How the messenger got its tail: addition of polyŽA. in the nucleus, Trends Biochem. Sci. 15 Ž1990. 277–281. M. Yokoyama, R.S. Morrison, I.B. Black, C.F. Dreyfus, Septal neuron cholinergic and GABAergic functions: differential regulation by basic fibroblast growth factor and epidermal growth factor, Dev. Brain Res. 78 Ž1994. 201–209. J.P. Zajicek, M. Wing, N.J. Scolding, D.A.S. Compston, Interactions between oligodendrocytes and microglia: major role for complement and tumor necrosis factor in oligodendrocyte adherence and killing, Brain 115 Ž1992. 1611–1631.
Research report
Gene expression in activated brain microglia: identification of a proteinase inhibitor that increases microglial cell number Smita Thakker-Varia a
a,)
, Stella Elkabes a , Charles Schick b, Gary A. Silverman b, Lang Peng a , Ann C. Sherwood a , Ira B. Black a
Department of Neuroscience and Cell Biology, UniÕersity of Medicine and Dentistry of New Jersey-Robert Wood Johnson Medical School, Piscataway, NJ 08854, USA b Joint Program in Neonatology, Department of Pediatrics, HarÕard Medical School, Children’s Hospital, Boston, MA 02115-5737, USA Accepted 27 January 1998
Abstract Microglia, the intrinsic immune cells of the central nervous system, are activated in a variety of inflammatory brain diseases in which they play a pathogenetic role. However, mechanisms underlying activation are largely unknown. To begin elucidating molecular mechanisms associated with activation, we characterized the pattern of gene expression in virtually pure dissociated microglial cultures, using RT-PCR differential display. Microglia were activated with bacterial lipopolysaccharide ŽLPS., a traditional stimulant, and the profile of gene expression was compared to that in basal, control cultures. Activation resulted in altered expression of six genes. The cDNAs were isolated, sequenced and characterized. Homology searches identified three novel genes, and two that exhibited very high sequence similarity to the gene encoding squamous cell carcinoma antigen ŽSCCA.. SCCA Ž1 and 2. are tandemly arranged genes that encode two serine proteinase inhibitors Žserpins.. SCCA has been detected exclusively in cancer cells, and is a plasma marker for squamous cell carcinoma. Immunoblot analysis indicated that gene expression was accompanied by a 5-fold increase in the synthesis of SCCA protein in LPS-activated microglia. To assess potential biological actions of the SCCA serpins, SCCA1 protein was added to cultures. SCCA1 altered microglial morphology, and elicited a dramatic, 5-fold increase in cell number within 72 h. The effects appeared to be cell-specific, since the protein had no effect on other cell types: cortical astrocytes and neurons from cortex or basal forebrain were unaffected. We tentatively conclude that SCCA1 may play a cell-specific role in increasing cell number, a critical early step in microglial activation and brain inflammation. More generally, differential display of genes in the microglial model system may help define patterns of expression associated with CNS disease, thereby identifying pathogenetic mechanisms and new therapeutic targets. q 1998 Elsevier Science B.V. Keywords: Microglial activation; Lipopolysaccharides; Differential display; Squamous cell carcinoma antigen and serine proteinase inhibitor
1. Introduction In the central nervous system ŽCNS., response to injury and disease involves glial activation. After injury, microglial activation, in particular, takes place in a stereotypic pattern. Activation is characterized by proliferation and migration of microglia to the site of injury, accompanied by characteristic morphological, immunophenotypical and functional changes w6,9,21,26x. Activation may exert
) Corresponding author. UMDNJ-Robert Wood Johnson Medical School, 679 Hoes Lane, Piscataway, NJ 08854-5635. Fax: q1-732-2354990; E-mail: [email protected]
0169-328Xr98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 9 - 3 2 8 X Ž 9 8 . 0 0 0 3 4 - 5
both beneficial andror deleterious effects. Beneficial responses include tissue repair by secretion of growth factors, destruction of invading microorganisms and phagocytosis of potentially damaging debris. However, activated microglia also release potentially cytotoxic substances, such as free radical intermediates, NO, proteases and cytokines w8,10,28,33x. Despite these responses of microglia to CNS injury, little is known about the mechanisms underlying activation. To understand the molecular genetics of microglial activation, it is important to determine associated patterns of gene expression. Delineation of gene expression governing activation may help define regulation of the components of activation that mediate beneficial and deleterious actions of microglia.
100
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
To begin assessing the molecular basis of microglial activation, we employed a model of virtually pure dissociated cell cultures of microglia. Bacterial lipopolysaccharide ŽLPS., a component of the cell wall of gram negative bacteria, and a classic microglial stimulator, was used to activate the cultures. Activation of microglial cultures with LPS modifies cellular functions, including release of cytokines, proteases and reactive oxygen intermediates w8,10,28,33x. To examine the profile of gene expression that regulates activation, we employed RT-PCR differential display. This technique identifies differentially expressed genes by comparing steady-state levels of mRNAs in cells under various conditions, independent of RNA prevalence. The entire spectrum of both abundant and rare mRNAs is screened using this method. We report that during microglial activation, a number of known and unknown genes are differentially regulated. We identified squamous cell carcinoma antigen ŽSCCA. partial cDNA that was specifically induced in activated microglia and was associated with an increase in SCCA protein. The SCCA family consists of two nearly identical proteins, SCCA1 and SCCA2 w27x. SCCA1 exhibited cell-specific biologic effects. Increased microglial cell number with altered morphology was evident after treatment with purified protein. Thus, using this microglial culture model of activation in conjunction with the PCR-based method, we have begun to define genes and their functions associated with microglial activation. Identification of activation-associated genes may lead to discovery of new physiologic and pathogenic mechanisms involved in brain inflammation, and development of new therapeutic targets.
2. Materials and methods 2.1. Isolation of microglia Microglia were isolated from postnatal day 2 Sprague– Dawley rats ŽHilltop, Scottsdale, PA.. Cortices were dissected. Special care was taken to remove all meninges and blood vessels during dissection to minimize contamination by blood monocytes and macrophages. Mixed glial cultures were grown by the modification of the method of McCarthy and DeVellis w20x and shaken for 2 h at 180 rpm at 378C. Microglia were isolated by differential adhesion on plastic dishes Ž; 10 min, 378C., followed by trypsinization. Microglial cultures were grown in N2 medium containing 10% heat-inactivated fetal bovine serum. The microglial cultures obtained were ) 98% pure, 2 and 5 days after plating. The purity and identification of the cultures was assessed immunocytochemically using the CD11brc ŽOX-42. and ED1 markers w4,5,11,12,23x.
2.2. Isolation of astrocytes To establish astrocyte cultures, postnatal day 2 Sprague–Dawley rats ŽHilltop. were used. Mixed glial cells were grown as described above. Oligodendrocytes and microglia were removed by shaking at 250 rpm overnight. The adhering astrocytes were dissociated and plated in Eagle’s minimum essential medium containing 15% heat-inactivated fetal calf serum w32x. The astrocytes were identified and confirmed for purity by immunostaining for glial fibrillary acidic protein ŽGFAP.. Oligodendrocytes and O-2A progenitor cells were identified using antibodies to myelin basic protein ŽMBP. and A2B5, respectively. 2.3. Isolation of neurons Pregnant Sprague–Dawley rats ŽHilltop. were used. Embryos were recovered at embryonic day 17 ŽE17.. The dissections were performed as previously described w15x but with some modifications. Cells were plated at a density of 700 000 cellsr35 mm dishes coated with poly-D-lysine. The cultures were maintained in serum-free fully defined medium w32x. This medium restricts the growth of glia and promotes virtually pure neuronal cultures. 2.4. RNA isolation and differential mRNA display Total RNA from pure microglial cultures was extracted by the guanidine isothiocyanate method w2x. After removal of chromosomal DNA contamination in total RNAs, differential mRNA display analysis was carried out as described by Liang and Pardee w19x using the RNA image kit as instructed by the manufacturer ŽGene Hunter, MA.. Control reactions were performed in the absence of reverse transcriptase. Total RNA Ž2 m g. was reverse transcribed in a 20 m l reaction mixture using a one-base anchored primer HT11 A. The cDNAs were then amplified by PCR in the presence of Ž 35 S. dATP, and in controls, water was substituted for cDNA. Each set of the reaction mixtures included an arbitrary 13-mer Ž5X AAGCTTGATTGCC 3X , 5X AAGCTTCGACTGT 3X or 5X AAGCTTTGGTCAG 3X . from the RNAimage kit as 5X primer and HT11 A as 3X primer. The conditions for PCR were as follows: denaturation at 948C for 30 s, annealing at 408C for 2 min, and extension at 728C for 30 s; reactions were subjected to 40 cycles. Radiolabeled PCR products were analyzed by denaturing polyacrylamide gel electrophoresis. The gels were dried on Whatman paper without fixing. Putative differentially expressed bands were excised, resuspended in 100 m l of H 2 O for 10 min at RT, boiled for 15 min followed by reamplification in the absence of isotope in a 40-cycle PCR w18,19x. Reamplified cDNAs ranging from 100 to 300 bp were used for cloning and as templates for random priming.
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
101
2.5. TA cloning and sequence analysis
2.9. Cell counts
Reamplified cDNA fragments were cloned directly into the plasmid vector pCR II using the TA cloning kit ŽInvitrogen, CA.. Sequences of the cDNAs were acquired on an Applied Biosystems 373A automatic sequencer as per manufacturer’s specifications ŽApplied Biosystems, Foster City, CA.. Comparisons of the unknown nucleotide sequences to known sequences in the Genbank and EMBL data bases was performed using BLAST program ŽGenetics Computer, Madison, WI..
To assess the total number of cells per 35 mm dish, three randomly selected, non-overlapping 4.75 mm2 strips were counted Ž1.48% of the culture dish area..
2.6. Antibodies
3. Results
OX-42 and ED1 monoclonal antibodies were purchased from Serotec ŽOxford, UK.. The rabbit polyclonal antibody to GST-SCCA Žglutathione S-transferase. was characterized and successfully used for western blot analysis of pure proteins w24x, cell lines expressing SCCA1 and SCCA2 and tissue extracts Žpersonal communication.. 2.7. Western blot analysis Microglial cells plated in equal numbers were solubilized in lysis buffer ŽPBS–1% Triton X-100.. Cell lysates were incubated for 10 min at 48C. The protein content was determined with the Coomassie plus protein assay kit ŽPierce Chemical.. Samples containing equal amounts of protein concentration were denatured in Laemmli’s sample buffer Žcontaining b-mercaptoethanol. for 5 min and subjected to 6–12% gradient SDS-PAGE w17x. The proteins were transferred to polyvinylidenedifluoride ŽPVDF. membranes ŽMillipore. which were blocked for 1 h with a 4% solution of dry milk powder in 0.2% Tween 20–PBS. The PVDF membranes were incubated at 48C with rabbit polyclonal antisera against GST-SCCA Ž1:2000 diln. w24x for 1 h at RT or overnight at 48C. Membranes were washed, followed by 1 h incubation with HRP-conjugated IgG Ž1:500. at RT. The immunopositive bands were visualized by chemiluminescence using the ECL detection kit ŽAmersham, Arlington Heights, IL.. 2.8. Cell treatments Cells were grown in the absence or presence of the physiological dose of 100 ngrml LPS ŽSigma, St Louis, MI.. Cells were then harvested and used for RNA isolation or cell lysate preparation. Microglial cells, cortical astrocytes and neurons were incubated with 4.8 m grml of purified GST-SCCA1 Žglutathione S-transferase. protein for up to 5 days. The GST-SCCA1 fusion protein used was batch purified using glutathione–Sepharose 4B beads and was free of any bacterial contaminants. The cells were observed and monitored on daily basis. To confirm the specificity of SCCA1 effects, microglia were incubated with purified GST for 5 days, as controls.
2.10. Statistical analysis Data were analyzed by Student’s t test or one-way ANOVA.
3.1. Microglial culture model system To characterize the profile of gene expression associated with activation, we examined virtually pure microglial cultures. We used CD11brc ŽOX-42. and ED-1, specific markers for the microglialrmacrophage lineage, to define purity. Ninety-eight to ninety-nine percent of the cells expressed ED-1, and CD11brc ŽOX-42.. One to two percent of the cells were A2B5-positive, O-2A progenitor cells. No GFAP-positive cells were detected. The cultures were exposed to LPS Ž100 ngrml., for 24 h. The lowest LPS concentration that elicited changes in microglial trophic factor and receptor expression was chosen for the present studies Žunpublished results.. 3.2. Differential mRNA display To identify differentially expressed genes potentially involved in microglial activation, we compared patterns of mRNA expression in control cultures with those stimulated by LPS for 24 h. Three different combinations of primer sets were used to amplify subsets of mRNA ŽFig. 1.. The primers consisted of 1 anchored primer HT11 A, and three short arbitrary primers w5X AAGCTTGATTGCC 3X , Žlanes A and B., 5X AAGCTTCGACTGT 3X Žlanes C and D., or 5X AAGCTTTGGTCAG 3X Žlanes E and F.x. Amplified cDNA species in control and activated microglia were largely identical, indicating that the general gene expression profile remained unaltered. Nevertheless, specific differences were apparent with activation. Lanes A, C and E represent products from control cultures, and lanes B, D and F represent PCR products from LPSactivated cells ŽFig. 1.. Differential expression was observed in a number of cDNAs. We concentrated on qualitative differences; six representative bands were chosen for further analysis. These PCR products were prominent bands Žindicated by arrows in Fig. 1. and exhibited differential expression during LPS-activation. Activation was associated with both increased and decreased gene expression. The cDNAs from bands 2–6 exhibited differential upregulation of mRNA during LPSactivation: expression was detectable only in LPS-activated
102
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
samples and not in controls. The cDNA from band 1 reflected marked downregulation of mRNA during LPSactivation; mRNA expression was detectable in control samples, and was undetectable after activation. To validate the display results, the bands were recovered from the dried gel and reamplified using corresponding primer sets. The cDNA fragment from band 5 did not reamplify and was not pursued further. ŽFailure to reamplify this cDNA could have resulted from poor recovery of the cDNA from the gel or from ‘false positivity’.. cDNA products were cloned into the PCR II plasmid vector. The nucleotide sequence of these cDNAs was determined. At least five recombinant subclones for each selected band were systematically sequenced Ždata not shown.. All the cDNAs appeared to be AT rich, indicating that they corresponded to the 3X end of the mRNAs. Some of the cDNAs contained a putative polyadenylation signal w31x proximate to the upstream sequences complementary to the anchored oligo dT primers. Consequently, these cDNA fragments correspond to the 3X end of the mRNAs, as expected. 3.3. Sequence analysis and homology Homology searches revealed that two of the activated sequences corresponded to known genes, validating the display approach ŽTable 1.. The 196 and 200 bp fragments Žbands 2 and 3. corresponded to bases 1609–1707 Žwith 69% homology. of the published sequence of human squamous cell carcinoma antigen cDNA Žaccession no. S66896. w27x including the 3X untranslated sequence. SCCA is a member of the serine proteinase inhibitor superfamily. No known genes with significant homology to bands 1, 4, and 6 Ž176, 97 and 167 bp. were found in the Genbank or EMBL DNA data bases. 3.4. Expression of mature SCCA protein To begin assessing the biological significance of differential gene expression, we initially focused on SCCA. Western blot analysis was performed to determine whether the transcribed message for SCCA induced during activation was translated into mature protein. The expression of SCCA protein was markedly increased in LPS-activated microglial cell lysates, corroborating the results obtained
Table 1 Analysis of cDNA fragments identified by differential mRNA display Fig. 1. Differential display comparing RNAs from control and LPS-treated microglia cells. Total RNA was extracted from control Žlanes A, C, E. and LPS-treated microglial culture Žlanes B, D, F. and subjected to differential mRNA analysis. Autoradiograms of amplified PCR products are shown for three different primer combinations that identified distinct fragments Žarrows. with differential expression in the LPS-activated X group. Primer combinations included HT11 A as 3 primer for all reactions X X X and three arbitrary 13-mers Ž5 AAGCTTGATTGCC 3 , 5 AAGCTTCX X X X . GACTGT 3 or 5 AAGCTTTGGTCAG 3 as 5 primer.
Band number
Fragment size
Regulation
Sequence homology
1 2
176 bp 196 bp
down up
3
200 bp
up
4 6
97 bp 167 bp
up up
no homology human squamous cell carcinoma antigen human squamous cell carcinoma antigen no homology no homology
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
103
Fig. 2. ŽA. Immunoblot analysis of SCCA protein in control and LPS-treated microglia cells. Equal amounts Ž50 m g. of protein were loaded in each lane, electrophoresed and immunoblotted with rabbit polyclonal antisera specific for GST-SCCA. Potential SCCA–proteinase complex Ž; 72 kDa. and SCCA Ž; 42 kDa. bands Žindicated by arrows. show increased expression of both proteins in LPS-treated cells. Panel ŽB. shows the densitometric quantitation of the ; 42 kDa band. The results are expressed as mean "S.E.M. of three independent experiments Ž p - 0.006 by Student’s t test..
by differential display ŽFig. 2A.. Control samples were generated by incubating microglia in N2 medium containing 10% heat-inactivated fetal bovine serum, but no LPS. SCCA-specific protein of ; 42 kDa was detected in samples treated with LPS for 24 h Žlane 2.. The polyclonal antibody to the GST-SCCA conjugate also recognized a band migrating at ; 72 kDa w7,24x, corresponding in molecular mass to endogenous SCCA–proteinase complex. Serpins form SDS-stable complexes at a stoichiometry of 1:1, upon binding to their target proteinases w24x. Densitometric analyses ŽFig. 2B. of the samples from three different experiments are expressed as percent of control. A 5-fold increase of SCCA-specific protein in the LPS-treated cells was evident. Control cells exhibited barely detectable levels of SCCA Žlane 1.. Thus, both the mRNA for SCCA and its protein product were induced in microglia with activation by LPS.
was observed. This range of increases may be due to cell culture variation. No change was observed in control cultures treated with GST alone Žnot shown.. In addition to increased cell number, SCCA1-treated microglia exhibited marked morphological changes, consistent with activation. SCCA1 elicited a marked increase in microglial cell size. Cells assumed a flattened, circular
3.5. Biological effects of SCCA1 To assess the potential biological effects of SCCA family serpins Žconsisting of SCCA1 and SCCA2., on microglia, cells were treated with purified GST-SCCA1 or control GST protein. After 72 h of treatment with purified SCCA1, a dramatic, 5-fold increase in cell number was evident ŽFig. 3.. In subsequent experiments using different culture preparations, a 2- to 5-fold increase in cell number
Fig. 3. Effect of SCCA1 on microglial cell number. Sister cultures in triplicate were plated in 35 mm dishes, 4.8 m grml of purified GSTSCCA1 was added to the cultures after 3 h and the cell number was assessed 72 h after plating. A 5-fold higher cell number was seen in the SCCA1-treated cultures. Values are the mean of triplicate dishes "S.E.M. in a representative experiment. The experiment was repeated 3 times. )Significantly different from control Ž p- 0.0001. by Student’s t test.
104
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
105
appearance, or an amoeboid configuration, with diameters of 40–60 m m, characteristic of activated microglia ŽFig. 4A.. In contrast, GST control cells were small and spindle-shaped, with only rare circular forms. Nevertheless, a degree of heterogeneity was evident with SCCA1 stimulation. Large process-bearing cells and smaller clubshaped cells were also evident. Finally, SCCA1 exposure was associated with cytoplasmic granularity, principally associated with the large, circular cells. In contrast, GSTtreated cultures resembled untreated controls ŽFig. 4B..
ential display screening method involves PCR amplification of subsets of mRNAs and separation of the resulting cDNAs by gel electrophoresis. This technique is extremely sensitive and identifies both abundant and rare mRNAs. The combination of virtually pure microglial culture with differential display appears to constitute a fruitful approach to elucidating the molecular genetics of activation and inflammation.
3.6. Cell specificity of SCCA1 effects
Our initial screen identified 6 differentially regulated cDNAs after LPS stimulation, including both novel and previously defined genes. Out of six cDNAs, five were upregulated and one cDNA was downregulated. These cDNAs with altered expression may represent genes important for one or more of the functional components of activation. Further characterization of the novel genes identified in our study will clarify their potential roles. In this study we have focused on a known gene of interest, SCCA, and its relation to the activation process in pure microglial culture.
Since microglia synthesize and respond to SCCA, actions may represent autocrine andror paracine regulation. To begin ascertaining the specificity of SCCA1 for microglia, biological effects on other cell types were evaluated. We compared effects on astrocytes, basal forebrain neurons and cortical neurons. SCCA1 exhibited no effects on cell number or morphology in cultures of cortical astrocytes, or neurons from cortex or basal forebrain. After 48 h of treatment with SCCA1, no significant change in cortical astrocyte number was observed Žcontrol:1.36= 10 5 " 0.33; SCCA1 treated: 1.43 = 10 5 " 0.88.. Similarly, after 72 h of treatment with SCCA1, neurons from either basal forebrain Žcontrol: 1.0 = 10 5 " 2.03; SCCA1 treated: 1.2 = 10 5 " 0.88. or cortex Žcontrol: 5.4 = 10 5 " 2.92; SCCA1 treated: 6.2 = 10 5 " 0.67. exhibited no significant change in number. The morphology of the neurons was unaltered. Thus SCCA1 exerts cell-specific effects on microglia, not evident in other cell types.
4. Discussion 4.1. Microglial actiÕation alters the gene expression profile Microglial activation is a hallmark of diverse CNS disorders. Activated microglia are characterized by multiple morphological and functional changes, including chemotaxis, proliferation, phagocytosis and cytoskeletal rearrangements. The signaling mechanisms regulating these changes in microglia and the associated gene expression are largely undefined. To identify genes potentially involved in microglial activation, we used a potent stimulator, LPS, to activate virtually pure dissociated cultures of microglia. Differential mRNA display was used to delineate gene expression associated with activation. The differ-
4.2. OÕerall profile
4.3. Identification of SCCA Within 24 h of LPS treatment, a number of genes were upregulated. We identified a gene of the serine proteinase inhibitor family, squamous cell carcinoma antigen. The upregulation of this serpin Žsee below. was measurable within 24 h, a time course consistent with an inflammatory response. Furthermore, elevation of SCCA message was associated with a 5-fold increase in SCCA protein expression. This observation simultaneously served to validate the display approach, and indicate that increased SCCA gene expression is accompanied by elevation of the protein product. Activation of microglia by LPS, therefore, apparently increased the levels of SCCA by both transcriptional and translational mechanisms. Elevation of SCCA protein itself raised the possibility of biologically relevant effector activities. 4.4. SCCA protein increases microglial number To define potential functions for SCCA in the inflammatory response, specific biological effects of one of the two serpins, SCCA1 were examined. Purified SCCA1 protein induced a 5-fold increase in microglial cell number, suggesting possible autocrine or paracrine function. During inflammation and injury, microglial infiltration associated with increased cell number is observed at the
Fig. 4. Effect of SCCA1 on microglial morphology. Microglial cells were incubated with 4.8 m grml of purified GST-SCCA1 protein for up to 5 days and controls were incubated with buffer. Cells were monitored on a daily basis and counted after 72 h. In the presence of SCCA1, the cells exhibited altered morphology typical of activation with an increased cell number ŽB.. Arrows indicate circular amoeboid cells and arrowhead indicates large process-bearing cell. Typical smaller spindle-shaped morphology was seen in control cells ŽA. shown by undulating arrows. Scale bar equals 40 m m.
106
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107
affected site. The increase may result from either chemotaxis, increased proliferation, andror enhanced survival. In our studies, increased cell number may have been attributable to increased mitosis, enhanced survival, or both w3x. We are presently examining these alternatives experimentally w3x. LPS-induced secretion of SCCA may enhance survival or enhance proliferation by inhibiting one or more serine proteinases inhibitory to microglia. 4.5. SCCA induces altered morphology Microglial cell morphology was significantly altered by purified SCCA1. SCCA1-treated cells were characterized by a large cell body with flattened, circular amoeboid-like morphology, consistent with activation. Conversion of resting microglia to an activated form occurs in response to injury or inflammation accompanied by characteristic changes in morphology. Proteases are essential to cytoskeletal rearrangement and cell adhesion properties associated with morphological restructuring. Increased microglial SCCA during inflammation may modulate these effects.
astrocytes, PN-1 is upregulated after transient ischemia w13x. PN-1 is also expressed in brain regions where continuous remodeling of neuronal connections takes place. Another serpin, a-1 antichymotrypsin is present in the Alzheimer amyloid plaques w30x and may facilitate deposition of b-amyloid peptide. By analogy with other serpins, SCCA1 may be involved in activating microglia by inducing proliferation. Increased expression of SCCA by activated microglia may also have a chemotactic function. The serpin-enzyme receptor complex found on surfaces of monocytes and neutrophils w14x may also be present on microglia, and SCCA might mediate neutrophil and or microglial chemotactic activity at injury sites. In conclusion, using differential display and an in vitro model of pure dissociated microglia, we have identified both known and novel genes that exhibit differential expression during activation. Among the known genes, SCCA, a member of the serpin family alters cell number.
Acknowledgements 4.6. SCCA1 effects are cell-specific The specificity of the effects of SCCA1 was examined using cortical astrocytes, as well as neurons from cortex and basal forebrain. In marked contrast to effects on microglia, SCCA1 did not alter cell number of the astrocytes or cortical or basal forebrain neurons. Moreover, this serine proteinase inhibitor did not alter cellular morphology of these cell types. Taken together, our findings of increased microglial number in the presence of SCCA1 and specific alteration in morphology, demonstrate that SCCA1 specifically and selectively affected microglia. SCCA is a member of the ovalbumin family of serine proteinase inhibitors. Recent studies have indicated that two nearly identical, tandemly arrayed SCCA genes are found in the genome w25x. Although the amino acid sequences of SCCA1 and SCCA2 are nearly identical, there are differences in reactive site loops suggesting differences in specificity for target proteinases. SCCA2 is effective against neutrophil cathepsin G and mast cell chymase w24x. Since inhibitory serpins regulate serine proteinases in diverse processes, including inflammation, cell migration and extracellular matrix remodeling w1,7,22,29x, SCCA may be an important regulator of increased proteolysis during microglial activation. SCCA serves as a serological marker for more advanced tumors of the cervix, lung and oropharynx w15,16x, but its functional role is not well defined either in normal or malignant cells. Although SCCA has not been previously associated with microglia, our results suggest that it may be an important regulator of microglial activation. A number of CNS protease inhibitors have been associated with disease processes, for example protease nexin-1 ŽPN-1. and a-1 antichymotrypsin. In reactive hippocampal
This work was supported by NISCD-35572-02 grant to S.E. and by NICHD-23315 and APA IBC-9501 grants to I.B.B. and by NIH CA 69331 to G.A.S.
References w1x R.W. Carrel, K.S. Aulak, M.C. Owen, The molecular pathology of the serpins, Mol. Biol. Med. 6 Ž1989. 35–42. w2x P. Chomczynski, N. Sacchi, Single step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extract, Anal. Biochem. 162 Ž1987. 156–169. w3x E. DiCicco-Bloom, W.J. Freidman, I.B. Black, NT-3 stimulates sympathetic neuroblast proliferation by promoting precursor cells, Neuron 11 Ž1993. 1101–1111. w4x C.D. Dijkstra, E.A. Dopp, P. Joling, G. Kral, The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulations in the rat recognized by monoclonal antibodies ED1, ED2 and ED3, Immunology 54 Ž1985. 589–599. w5x S. Elkabes, E.M. DiCicco-Bloom, I.B. Black, Brain microgliarmacrophages express neurotrophins that selectively regulate microglial proliferation and function, J. Neurosci. 16 Ž1996. 2508–2521. w6x J. Fulcrand, A. Privat, Neurological reaction secondary to Wallerian degeneration in the optic nerve of the postnatal rat: ultrastructural and quantitative study, Proc. Natl. Acad. Sci. U.S.A. 176 Ž1977. 189–224. w7x P.G.W. Gettins, P.A. Patston, S.T. Olson, Serpins: Structure Function and Biology, Molecular Biology Intelligence Unit, R.G. LandesrChapman & Hall, Austin, TX, 1996. w8x D. Giulian, Microglia, cytokines and cytotoxins: modulators of cellular responses after injury to the central nervous system, J. Immunol. Immunopharmacol. 10 Ž1990. 15–21. w9x D. Giulian, J. Chen, J.E. Ingeman, J.K. George, M. Noponen, The role of mononuclear phagocytes in wound healing after traumatic injury to adult mammalian brain, J. Neurosci. 9 Ž1989. 4416–4429. w10x D. Giulian, K. Vaca, M. Corpuz, Brain glia release factors with opposing actions upon neuronal survival, J. Neurosci. 13 Ž1993. 29–37.
S. Thakker-Varia et al.r Molecular Brain Research 56 (1998) 99–107 w11x M.B. Graeber, W.J. Streit, G.W. Kreutzberg, Axotomy of the rat facial nerve leads to increased CR3 complement receptor expression by activated microglial cells, J. Neurosci. Res. 21 Ž1988. 18–24. w12x M.B. Graeber, W.J. Streit, G.W. Kreutzberg, The third glial cell type, the microglia: cellular markers of activation in situ, Acta Histochem. 37 Ž1990. 157–160, Suppl. w13x M.C. Hoffmann, C. Nitsch, A.L. Scotti, E. Reinhard, D. Monard, The prolonged presence of glia-derived nexin, an endogenous protease inhibitor, in the hippocampus after ischemia-induced delayed neuronal death, Neuroscience 49 Ž1992. 397–408. w14x G. Joslin, G. Griffin, A. August, S. Adams, R. Fallon, R. Senior, D. Perlmutter, The serpin-enzyme complex ŽSEC. receptor mediates the neutrophil chemotactic effect of a-antitrypsin-elastase complexes amyloid-b peptide, J. Clin. Invest. 90 Ž1992. 1150–1154. w15x H. Kato, in: S. Sell ŽEd.., Serological Cancer Markers, Humana Press, Totowa, NJ, 1992, pp. 437–451. w16x H. Kato, T. Torigoe, Radioimmunoassay for tumor antigen of human cervical squamous cell carcinoma, Cancer 40 Ž1977. 1621–1628. w17x U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4 , Nature 227 Ž1970. 680–685. w18x P. Liang, L. Averboukh, A.B. Pardee, Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction, Nucleic Acids Res. 21 Ž1993. 3269–3275. w19x P. Liang, A. Pardee, Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction, Science 257 Ž1992. 967–971. w20x K.D. McCarthy, J. DeVellis, Preparation of separate astroglial and oligodendroglial cell cultures from rat cerebral tissue, J. Cell Biol. 85 Ž1980. 890–902. w21x Y. Murabe, Y. Ibata, Y. Sano, Morphological studies on neuroglia: III. Macrophage response and ‘microgliocytosis’ in kainic acid induced lesions, Cell Tissue Res. 218 Ž1981. 75–86. w22x J. Potempa, E. Korzus, J.T. Korzus, The serpin superfamily of proteinase inhibitors: structure, function, and regulation, J. Biol. Chem. 269 Ž1994. 15957–15960. w23x A.P. Robinson, T.M. White, D.W. Mason, Macrophage heterogeneity in the rat is delineated by two monoclonal antibodies MCR
w24x
w25x
w26x
w27x
w28x
w29x w30x
w31x w32x
w33x
107
Ox-41 and MRC Ox-42, the latter recognizing complement receptor type 3, Immunology 57 Ž1986. 239–247. C. Schick, Y. Kamach, A.J. Bartuski, S. Cataltepe, N.M. Schechter, P.A. Pemberton, G.A. Silverman Pemberton, Squamous cell carcinoma antigen 2 is a novel serpin that inhibits the chymotrypsin-like proteinases cathepsin G and mast cell chymase, J. Biol. Chem. 272 Ž1997. 1849–1855. S. Schneider, C. Schick, K. Fish, E. Miller, J. Pena, S. Treter, S. Hui, G. Silverman, A serine proteinase inhibitor locus at 18q21.3 contains a tandem duplication of the human squamous cell carcinoma antigen gene, Proc. Natl. Acad. Sci. U.S.A. 92 Ž1995. 3147– 3151. G. Stoll, B.D. Trapp, J.W. Griffin, Macrophage function during Wallerian degeneration in the rat optic nerve: clearance of degenerating myelin and Ia expression, J. Neurosci. 9 Ž1989. 2327–2335. Y. Suminami, F. Kishi, K. Sekiguchi, H. Kato, Squamous cell carcinoma antigen is a new member of the serine protease inhibitors, Biochem. Biophys. Res. Commun. 181 Ž1991. 51–58. C. Thery, B. Chamak, B. Mallat, Cytotoxic effect of brain macrophages on developing neurons, Eur. J. Neurosci. 3 Ž1991. 1155–1164. J. Travis, G.S. Salvesen, Human plasma proteinase inhibitors, Annu. Rev. Biochem. 52 Ž1983. 655–709. S. Wagner, J. Geddes, C. Cotman, A. Lau, D. Gurwitz, P. Isackson, D. Cunningham, Protease nexin-1, an antithrombin with neurite outgrowth activity, is reduced in Alzheimer disease, Proc. Natl. Acad. Sci. U.S.A. 86 Ž1989. 8284–8288. M. Wickens, How the messenger got its tail: addition of polyŽA. in the nucleus, Trends Biochem. Sci. 15 Ž1990. 277–281. M. Yokoyama, R.S. Morrison, I.B. Black, C.F. Dreyfus, Septal neuron cholinergic and GABAergic functions: differential regulation by basic fibroblast growth factor and epidermal growth factor, Dev. Brain Res. 78 Ž1994. 201–209. J.P. Zajicek, M. Wing, N.J. Scolding, D.A.S. Compston, Interactions between oligodendrocytes and microglia: major role for complement and tumor necrosis factor in oligodendrocyte adherence and killing, Brain 115 Ž1992. 1611–1631.