198      J. Mex. Chem. Soc. 2013, 57(3)  Mariana Poggi et al. New Isoniazid Complexes, Promising Agents Against Mycobacterium tuberculosis Mariana Poggi,1 Rafael Barroso,2 Antonio José Costa-Filho,2 Heloisa Barbosa de Barros,3 Fernando  Pavan,3 Clarice Queico Leite,3 Dinorah Gambino,*1 and María Helvecia Torre*1 1  Cátedra de Química Inorgánica, Facultad de Química, Universidad de la República, Gral Flores 2124, C. C. 1157, 11800  Montevideo, Uruguay 2  Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto, Universidade de São Paulo, 14040-901, Ribeirão Preto (SP),  Brazil 3  Departamento de Ciências Biológicas, Faculdade de Ciências Farmacêuticas, Universidade Estadual Paulista, UNESP,  Rodovia Araraquara Jaú Km, 01, 14801-902 Araraquara, SP, Brazil   mtorre@fq.edu.uy, dgambino@fq.edu.uy Received February 27, 2013; Acepted June 5, 2013. J. Mex. Chem. Soc. 2013, 57(3), 198-204 © 2013, Sociedad Química de México ISSN 1870-249X Article Abstract. Tuberculosis (TB) is a public health disease that produces  several million deaths annually worldwide. Due to this critical situa- tion and the appearance of drug-resistant microbial strains, innovation  in TB drug discovery is a research priority. In this work, the synthesis  and  characterization  by  elemental  analysis,  thermogravimetry,  con- ductimetric  measurements  and  spectroscopies  UV-Vis,  IR  and  EPR  of  [Cu(INH)(H2O)]SO4⋅2H2O  (Cu-INH)  and  [CoCl(INH)2(H2O)]  Cl⋅2.5H2O  (Co-INH)  complexes  with  isoniazid  (INH)  are  reported.  Besides,  the  lipophilicity,  the activity against Mycobacterium tuber- culosis (MICCu-INH = 0.78 µg/mL and MICCo-INH = 0.19 µg/mL) and  the cytotoxicity (IC50 = 48.8 and 625 µg/mL for  the copper and co- balt complexes, respectively) were measured and the selectivity index  (62.5 for Cu-INH and 3205 for Co-INH) was calculated. These results  indicate that these complexes are good candidates for further studies. Keywords: copper and cobalt complexes, isoniazid, antimycobacteria  activity, tuberculosis, spectroscopy. Resumen. La tuberculosis (TB) es una enfermedad pública que produ- ce varios millones de muertes anualmente en todo el mundo. Debido a  esta crítica situación y a la aparición de cepas resistentes a los fármacos  usuales,  la  investigación de nuevas moléculas para el  tratamiento de  la TB se ha vuelto una prioridad. En este trabajo se reporta la síntesis  y  caracterización  por  análisis  elemental,  termogravimetría,  medidas  conductimétricas y espectroscopías UV-Vis, IR y EPR de los comple- jos [Cu(INH)(H2O)]SO4⋅2H2O (Cu-INH) y [CoCl(INH)2(H2O)]Cl⋅2.5 H2O (Co-INH) con isoniacida (INH). Por otra parte, se determinó la  actividad anti-Mycobacterium tuberculosis (CIMCu-INH = 0.78 µg/mL  y CIMCo-INH = 0.19 µg/mL) y la citotoxicidad de los complejos (IC50  = 48.8 and 625 µg/mL respectivamente para los complejos de cobre  y cobalto) y se calculó el índice de selectividad (62.5 para Cu-INH y  3205 para Co-INH). Estos resultados indican que estos complejos son  buenos candidates para continuar estudios futuros. Palabras clave:  complejos de cobre y cobalto,  isoniacida, actividad  antimicobacteria, tuberculosis, espectroscopía. Introduction Tuberculosis  (TB)  is  a  public  health  disease  that  produces  several millions deaths annually worldwide [1]. In developing  countries is a leading cause of mortality. Nowadays, the number  of infections is growing also in developed countries, especially  in immunocompromised patients such as those co-infected with  human  immunodeficiency virus  (HIV),  and  in  individuals  re- ceiving anti-tumour therapy or diabetics [2-4]. A big associated problem is the alarming increase of drug- resistant microbial strains (MDR-TB)  that makes difficult  the  effective control of the disease. Besides, to further complicate  the matter, drug-drug interactions between TB drugs and anti- HIV  treatments  or  other  chronic  disease  medications  such  as  those used in diabetics are observed [2]. Due  to  this  critical  situation,  innovation  in  TB  drug  dis- covery and evolving  strategies  to bring new agents with best  performance is a current health priority. One strategy to obtain new antimycobacterial drugs is the  development of new molecules by reengineering old drugs with  the aim to improve the antimycobacteria activity and to obtain  better resistance profiles, bioavailability and tolerability, among  others [3]. In particular, the coordination of metal cations with  organic drugs is a promising strategy that has been successful in  many cases with different pharmacological activities [5-7]. In  particular, in the research of new antimycobacteria compounds  with best activity and/or lower resistant effects, several metallic  complexes were reported in the literature. For example, silver  complexes with α-hydroxycarboxilic acids have proved  to be  effective  antimycobacterial  compounds,  potential  candidates  for antiseptic or disinfectant products  [8]. On  the other hand,  several series of homoleptic and heteroleptic iron, copper and  ruthenium compounds were developed being more active than  the free  ligands against Mycobacteria tuberculosis (M. tuber- culosis) [9-14]. Taking  into  account  these  antecedents  and  as  a  part  of  our work  in  the  research of  new antimicrobial molecule  [15- 19],  in  this article  the development of new Cu(II)  and Co(II)  complexes  with  isoniazid  (INH)  is  reported  and  the  activity  against M. tuberculosis  is evaluated. Besides, the cytotoxicity  in mammalian cells was determined and  the selectivity  index  was calculated. Isoniazid (Figure 1), the first effective drug discovered for  the treatment of  tuberculosis [20],  is  included in the group of  the  first-line  antimycobacterial  drugs  used  in  prevention  and  treatment of tuberculosis. However, shortly after its discovery,  New Isoniazid Complexes, Promising Agents Against Mycobacterium tuberculosis  199 INH  resistant  M. tuberculosis  strains  were  reported  [21].  Al- though this, INH is one of the most used drugs today. INH is able to coordinate with metal cations through dif- ferent chemical groups: heterocyclic nitrogen from the pyridine  ring and/or carbonylic O and N atoms of the hydrazide group.  For  this  versatility  it  is  also  an  interesting  ligand  from  the  chemical point of view. Results and discussion Analytical results The elemental analyses of the complexes were: [Cu(INH)(H2O)]SO4⋅2H2O  (Code  Cu-INH),  Anal  C  21.2%, N 11.9%, H 3.4%, calcd for C6N3H13O8SCu, C 21.5%,  N 12.0%, H 3.7%; yield: 70%, 135mg. [CoCl(INH)2(H2O)]Cl⋅2,5H2O  (Code  Co-INH),  Anal  C  31.33%,  N17.21%,  H4.69%,  Cl  15.40%,  calcd  for  C12N6H21O5.5CoCl2,  C  30.82%,  N  17.98%,  H  4.49%,  Cl  15.20%; yield: 76%, 196 mg. These results agree with the proposed stoichiometries. Thermogravimetric and conductimetric measurements Thermogravimetric results are shown in Table 1. Experimental mass losses agree with calculated values for  the two complexes. The  thermal  process  corresponding  to  Cu-INH  firstly  shows the release of two hydration water molecules at 58.5 °C  while the release of the coordinated water molecule took place  at  112.9  ºC,  as  expected.  The  decomposition  of  the  organic  moiety starts above 140 ºC forming the copper oxide at the end  of calcination. Similarly, the thermogram of Co-INH indicates  that  the  complex  has  2.5  hydration  water  molecules  and  one  coordinated molecule. Results  of  the  conductimetric  measurements  in  dimethyl  sulfoxide (DMSO) solutions are shown in Table 2. According to previous reports the conductivities obtained  for both complexes are in agreement with the assigned formula  for a 1:1 electrolyte type in DMSO [22-23]. In  addition,  the  conductivity  was  measured  in  the  same  solutions  after  48  hs  and  no  major  changes  were  observed  showing the stability of the complexes in DMSO. Spectroscopic measurements Infrared spectra The IR spectra of the complexes were compared with those of  the free ligands and previously reported complexes [24-27]. Table 3 shows the tentative assignments of the main bands  of free isoniazid and the metal complexes. Analysis of Cu-INH spectrum According  to  bibliographic  reports  [24,  25],  the  INH  spec- trum presents two bands at 3300 and 3209 cm-1 corresponding  to asymmetric and symmetric stretching of NH2 group. Upon  copper complexation the νa(NH2) shifted to lower frequencies  while the νs(NH2) was not identified due to the broad bands in  the region [26]. Besides, the band assigned as ν(N-H) doesn’t shift in the complex showing that this group does not participate  in  the coordination. On  the other hand,  INH spectrum shows  one  band  at  1667  cm-1 assigned as ν(C=O). In the complex Table 1. Thermal data for the metal-INH complexes. Complex Temperature range (ºC) DrTGA minima (ºC) Mass loss (%) Found/Calc. Removed group Cu-INH 20-80 58.5 10.6/10.3 2 H2O 80-140 112.9 4.6/ 5.1 H2O Co-INH 20-80 40.4 9.2/9.6 2.5 H2O 80-130 100.3 3.9/3.8 H2O Table 2. Conductivity (µS) in DMSO at 25 °C. Complex ΛDMSO(µS) at 25°C Cu-INH 35.3 Co-INH 52.0 Table 3.  Main  IR  bands  and  their  assignment  for  Cu-INH  and  Co- INH. INH Cu-INH Co-INH Assignment — 3409 (br) 3378 (br) ν(O-H)water 3300 (m),  3209 (m) 3247 (m, br.),  — 3209 (s),  3176 (s) ν(NH2) 3114 (s) 3114 (s) 3114 (s) ν(N-H) 1667 (vs) 1655 (vs) 1650 (vs) ν(C=O) 1635(s),  1557(vs) 1648 (sh),  1543 (vs) 1637(sh),  1550 (s) ν(C-N)py 1492 (m) 1498 (m) 1500 (s) δ(CNH) 1334 (s) 1370 (m) 1333 (m) ν(C-N)amide 1142 (s) 1148 (s) 1140 (m) ν(N-N) 888 (m) 864 (m)| 898 (m) δ(C-N-C)py 504 (m) 570 (m) 535 (m) δ(C-C-C)py — 1118 (s) — ν(S-O) — 610 (s) — δ(S-O) — 281 (w) 279 (m) ν(M-N) s = strong, vs = very strong, m = medium, w = weak, sh = shoulder,  br = broad. Fig. 1. Structure of isoniazid. 200      J. Mex. Chem. Soc. 2013, 57(3)  Mariana Poggi et al. this band shifted to 1655 cm-1 remarking the copper coordina- tion through this group. Similar shifts were observed in com- plexes  where  the  metal  coordinate  through  the  –C=O  group  [27, 28]. This behavior permits  to propose that INH acts as a  bidentate  ligand coordinating  through  the –NH2 and  the C=O  groups. Regarding the ν(C-N)amide and ν(N-N) bands the coordina- tion affects only  the  first one. A plausible explanation would  be  that  the coordination  through  the C=O group weakens  the  C-O  bond  and  consequently  reinforces  the  N-H  one,  shifting  the ν(C-N)amide to a higher frequency. Upon complex formation, the shifts of pyridine vibrations  in the high-frequency region usually are small, whereas those  corresponding  to  bendings  are  more  evident  and  shifted  to  higher frequencies [26]. As Table 3 shows, in the high-frequen- cy region the band corresponding to νa(C-N)py  (1635  cm-1),  νs(C-N)py (1557 cm-1) and δ(C-N-C)py (888 cm-1) show minor  shifts  (13-24  cm-1)  to  higher  or  lower  frequencies.  On  the  contrary, the low-frequency region evidences more clearly the  coordination through the heterocyclic nitrogen atom. Especially  the  band  at  504  cm-1 assigned  as  a  ring  C-C-C  bending  [29]  shifted to 570 cm-1, confirming the coordination through pyri- dine. Due to the fact that the aromatic pyridine ring is rigid, the  coordination through the heterocyclic N atom should be given  with another copper ion different than that coordinated with the  hydroxamine group, in a dimeric or polymeric structure. Regarding the metal-ligand bands, only one band was ob- served at 281 cm-1 and it was assigned as ν(Cu-O) taking into account previous work [26, 29]. The  inorganic  sulfates  present  only  one  stretching  band  usually  found  in  the  range  1140-1080  cm-1.  Besides,  sulfate  bonds can bend giving rise to one or two bands in the 680-610  cm-1 range, sharper than the stretching bands [26]. In the case  of Cu-INH the bands at 1118 and 610 cm-1 confirm the pres- ence of a sulfate counterion. Besides,  the broad band at 3409 cm-1 confirms  the pres- ence of water molecules in the complex. Analysis of the Co-INH spectrum The Co-INH spectrum shows the shifts of νa(NH2) and νs(NH2)  to lower frequencies (see Table 3), according to the cobalt co- ordination through this group [26]. Besides, the ν(C=O) shifts to  lower frequencies  in a similar behavior  to  that observed in  Cu-INH spectrum. The ν(N-H) does not change in the complex showing that this group does not participate in the coordination.  Taking into account this analysis, we can propose that isoniazid  acts as a bidentate ligand. Unlike  the case of  the Cu-INH,  in  the Co-INH spectrum  no significant changes in the pyridine vibrations were observed.  This would discard  the  coordination  through  the heterocyclic  nitrogen atom. On the other hand, only one band was observed at 279 cm-1  and it was assigned as ν(Co-O) taking into account previous reports [26, 29]. Besides,  the broad band at 3378 cm-1 confirms  the pres- ence of water molecules in the complex. Electronic spectra Copper(II) complexes of lower coordination number can exist  in a wide range of stereochemistries, some of  them distorted,  making it difficult to use electronic spectroscopy for identifying  structures [30]. In the case of Cu-INH, the electronic spectrum  in a nujol suspension shows one poorly defined broad band at  600 nm almost overlapped by ligand bands. This result agrees  with  the  report of similar complexes with a CuN2O2 chromo- phore [30]. On  the  other  hand,  the  Co-INH  is  a  pale  pink  powder,  typically of octahedral  complexes  [31].  It  is known  that high  spin six coordinate cobalt(II) complexes generally show,  in a  visible  region,  a multiple  structured band assigned  to  4T1g(F)  → 4T1g(P)  [30].  In  the  reflectance spectrum of Co-INH  three  bands at 606, 695 and 725 nm, and a shoulder at 668 nm are  observed. EPR spectra The EPR spectrum of  a polycrystalline  sample of Cu-INH at  X-band is shown in Figure 2. This spectrum shows two main resonances centered around  290 and 325 mT and that are attributed to transitions allowed by  the interaction between the magnetic moment of the unpaired  electron and the spectrometer magnetic field (Zeeman interac- tion).  Those  two  resonances  are  characteristic  of  an  axially  symmetric paramagnetic  center with principal  components of  the g-tensor given by g⊥ = 2.023 and g// = 2.236 as calculated  directly  from  the  resonance  positions  (in  units  of  magnetic  field) and from the frequency of microwave radiation used in  the experiment. The relation g// > g⊥ indicates that the unpaired  electron  is  in  a  dx 2 -y 2  ground  state  orbital.  Copper  ions  have  nuclear  spin  of  3/2  that  would  give  rise  to  extra  resonance  lines  in  their  spectrum  (hyperfine  interaction).  However,  no  such  lines are observed  in Figure 2, which  is a situation usu- ally found in the cases where copper - copper interactions are  observed [32-36]. Fig. 2.  X-band  EPR  spectrum  of  a  polycrystalline  sample  of  Cu- INH. New Isoniazid Complexes, Promising Agents Against Mycobacterium tuberculosis  201 It  is  also  worth  noting  the  absence  of  extra  resonances  for both low (below 275 mT) and high (above 375 mT) fields  and also of the so-called half-field transition, which is typical  of  copper pairs  [37]. Together  these  features  suggest  that  the  copper  ions  in Cu-INH are arranged as a polymeric  structure  in the solid state, as shown in Figure 3. As a result of the experimental analytical data and the spec- troscopic studies (IR, UV-Vis and EPR) the proposed structures  for the complexes are shown in Figure 3. Lipophilicity and solubility Before the study of lipophilicity, the stability of complexes in  physiological  solution  was  determined.  A  thin-layer  chroma- tography  using  the  experimental  conditions  presented  in  3.4  was performed. The solutions analyzed were: saturated solution  of  Cu-INH  in  physiological  serum,  saturated  solution  of  Cu- INH  in  water,  saturated  solution  of  Co-INH  in  physiological  serum  and  saturated  solution  of  Co-INH  in  water.  After  the  experiment, only one spot was observed in each Co-INH chro- matogram and the Rf were 0.73 in both solutions, showing that  the presence of Cl- does not affect the species in the physiologi- cal  solution.  The  same  behavior  was  observed  in  the  case  of  Cu-INH but the Rf was almost 0 according to the low solubility  of the complex in methanol. With  the aim  to determine  the  lipophilic character of  the  complexes  that  can  be  correlated  with  biological  properties  like  the passage  through  lipophilic barriers, distribution coef- ficients (D) between n-octanol and physiological solutions were  determined.  Besides,  the  solubility  in  water  and  DMSO  was  estimated. These results are presented in Table 4. The D values obtained for the complexes are higher than  that of the free ligand. According to the results shown in Table  4, the compound most soluble in water (isoniazid) was the less  lipophilic while both complexes presented lower solubility and  were more lipophilic than isoniazid, as expected. The solubilities  in DMSO are higher  than  the values ob- tained in aqueous solution. For this reason this solvent was used  in the microbiological assays. Anti-M. tuberculosis activity and cytotoxicity on mammalian cells The  minimum  inhibitory  concentration  (MIC)  values  against  M. tuberculosis, the IC50 and the selectivity index of the metal- INH complexes and  the  free  ligand are presented  in Table 5.  None of the metallic salt dilutions (CuSO4 and CoCl2), used as  control, inhibited the growth of mycobacterium. In the case of  Cu-INH, the molecular weight of the monomeric complex was  used  to calculate  the MIC, due  to  the uncertainty  in knowing  the molecular weight in solution. If polymeric structures exist  in solution the calculated MIC values expressed in µM will be  lower. As table 5 shows, both complexes were active against M. tuberculosis.  The  MIC  values  are  lower  than  that  of  the  eth- ambutol (MIC = 5.62 μg/mL, 27.5 μM) and p-aminosalicylic (MIC = 1.25 μg/mL, 8.16 μM), antimycobacterial agents in clinical use. Besides, Cu-INH complex (MIC = 2.2 μM) is less active than isoniazid (MIC = 0.063μg/mL, 0.46 μM) while Co-INH (MIC = 0.41 μM) has similar activity but both complexes pre- sented  MIC  values  well  below  25  µg/mL,  considered  as  a  limit  to  continue  the  anti-mycobacterial  studies,  according  to  pipeline by Pavan et al.[38]. One  thing  to  note  is  that  Co-INH  was  more  active  than  Cu-INH and more lipophilic. However, isoniazid was less lipo- philic than the complexes but it was the most active one. INH is  a prodrug which is active after transformation, by a peroxidase,  to  isonicotinic  acid  that  inhibits  the  synthesis  of  the  mycolic  acid required for the biosynthesis of mycobacterial cell wall [9].  INH enters into the cell by passive transportation [39] [40]. In  spite of  its polarity  this molecule  is a very powerful drug.  In  the case of our complexes, two mechanisms would be possible:  Table 4. Distribution coefficients (D) between n-octanol and physio- logical solutions, and estimated solubility (S) of the complexes. Complex D/ logD* S (mg/100 mL water) S (mg/100 mL DMSO) INH 0.07/-1.2 1667 815 Cu-INH 0.16/-0.79** 3.0 5.8 Co-INH 0.33/-0.48** 75 150 * strength force = 0.15 M, 25 ºC. ** pH = 4.98 (for Cu-INH) and 4.38 (for Co-INH). Table 5. MIC (mg/mL), IC50 and Selectivity Index for Cu-INH, Co- INH. Compound MIC* (µg/mL) MIC* (µM) IC50(μg /mL)/IC50(μM) SI** VERO J774A.1 Cu-INH 0.78 2.2 62.5/178 58.6/167 80.1 Co-INH 0.19 0.41 >500/1070 >500/1070 2631.6 * The obtained MIC values were the same in the triplicate assays **The selectivity index (SI) was calculated by the ratio between  VERO/MIC.Fig. 3. Proposed structures for Cu-INH (A) and Co(INH) (B). 202      J. Mex. Chem. Soc. 2013, 57(3)  Mariana Poggi et al. the complexes would pass through the cell wall by a different  mechanism than isoniazid or they would act through a different  mechanism of action. Due to  the fact  that  the mode of action  of isoniazid requires the biotransformation to isonicotinic acid,  the stable coordination to a metal center would block the whole  process leading to an inactivation of isoniazid. In our case, both  new metal complexes are active allowing to conclude that the  mode of action of the complexes could be completely different  to  that  of  free  isoniazid.  If  this  hypothesis  could  be  verified  through further studies these compounds would be good drug  candidates. Another interesting aspect of these complexes is the cyto- toxicity shown on two different mammalian cells lineages. The  models chosen were the VERO (ATCC CCL81) and J774.A1  (ATCC TIB-67) cells. VERO cells are considered normal cells  and do not represent any disease. These cells are largely used  as phenotypic  screening  in drug discovery  in  response of  the  cytotoxicity of the compounds. J774A.1 are macrophage cells,  that  represent  the  first  ones  of  the  immunological  response.  Also, mycobacteria has the ability to survive inside them. All  the complexes showed moderate cytotoxicity on these cell lines  and good selectivity indexes (Table 5). As a conclusion, the synthesis and characterization of two  new isoniazid complexes are reported. According to the spec- troscopic  analysis  the  copper  complex  presents  a  polymeric  structure  where  the  isoniazid  acts  as  a  bridge  between  two  different  copper  ions,  coordinating  through  the  heterocyclic  nitrogen, the amine and carbonyl groups. The cobalt complex presents an octahedral geometry where  the  isoniazid  is a bidentate  ligand. These  two examples show  the versatility of isoniazid as ligand. Both  complexes  were  active  against  M. tuberculosis showed  MIC  values  well  below  25  µg/mL  considered  as  a  limit to continue the antimycobacterium studies. Besides, they  showed low cytotoxicity on different mammalian cell lines and  consequently they presented high selectivity indexes. Due to these results, Cu-INH and Co-INH are interesting  candidates for further studies. Experimental All  starting  materials  were  commercially  available  research- grade chemicals and were used without further purification. Synthesis The  copper  and  cobalt  complexes  were  synthesized  mixing  aqueous  solutions  of  CuSO4 .5H2O  (0.55  mmol,  137  mg)  or  CoCl2.6H2O (0.55 mmol, 130 mg) and isoniazid (1.10 mmol,  150  mg)  with  continuous  stirring  for  30  min.  The  resulting  green  or  pink  precipitates  corresponding  to  copper  or  cobalt  complexes respectively were filtered, washed with small por- tions of water and dried at room temperature. The elemental analysis was performed with a Carlo Erba  EA1108  elemental  analyser.  Sulphate  was  identified  by  pre- cipitation  with  BaCl2  and  chloride  was  quantified  through  a  potentiometric method using AgNO3. Thermogravimetric and conductimetric measurements Thermogravimetric measurements were performed with a Shi- madzu TGA 50 thermobalance, with a platinum cell, working  under  flowing air  (50 mL min-1)  and at a heating  rate of 0.5  °C min-1 until 80 ºC and 1 °C min-1 above 80 ºC. Conductimetric measurements using a Conductivity Meter  4310  Jenway  were  performed  at  25  °C  in  a  DMSO  solution  (10-4 M), solvent used in the microbiological assays. The sta- bility of the complexes in this solvent was followed by measur- ing conductivity of these solutions for 48 h of the dissolution  and compared with the initial conductivity. Spectroscopic measurements IR  spectra,  in  the  range  between  4000  and  200  cm-1,  were  recorded on  a Bomem M 102 FTIR  spectrophotometer using  the KBr pellet technique. Electronic  spectra  of  complexes  in  a  Nujol  suspension  were registered on a Shimadzu UV-1603 spectrophotometer. Room temperature EPR measurements of the copper com- plex were carried out on polycrystalline samples using a JEOL  JES-FA200 spectrometer. The experimental parameters were:  microwave  frequency,  9.4  GHz;  microwave  power,  10  mW;  field modulation, EPR  measurements  of  the  copper  complex  were  carried  out on polycrystalline samples using an X-band Varian E109  spectrometer.  The  experimental  parameters  were  microwave  frequency: 9.4 GHz, microwave power: 10 mW, field modula- tion: 100 kHz, modulation amplitude: 0.2 mT,  time constant:  0.03  s,  scan  time:  2  min.  The  parameters  were  optimized  to  avoid saturation and/or distortions Thin-layer chromatography With the aim of studying the stability of complexes in physi- ological  solution  and  in  water,  a  thin-layer  chromatography  was performed with a saturated solution of complexes in both  solvents. A silica gel G, 250 μm plate and a CH3OH:  NH3  (100:1.5) solution as a mobile phase were used. 10 μL of each  solution  (saturated  solution  of  Cu-INH  and  Co-INH  in  physiological solution, and saturated solutions of Cu-INH and  Co-INH in water) were seeded in the plate. After the chroma- tography, the spots were detected with UV lamp [41]. Lipophilicity test and solubility Lipophilicity  tests  at  25  ºC  were  performed  determining  the  partition  coefficient  (P)  between  n-octanol  and  physiological  solution [42]. The concentration of the compounds in the physi- ological  solution  before  and  after  the  contact  with  n-octanol  was determined by UV spectroscopy at 260 nm. This isoniazid  band does not shift with complexation. New Isoniazid Complexes, Promising Agents Against Mycobacterium tuberculosis  203 The  solubility  at  37  °C  in  water  and  DMSO  was  also  estimated. Anti-M. tuberculosis activity The anti-MTB activity of  the  compounds was determined by  the  REMA  (Resazurin  Microtiter  Assay)  method  according  to  Palomino  et al,  2002  [43].  Stock  solutions  of  the  tested  compounds were prepared in dimethyl sulfoxide (DMSO) and  diluted in Middlebrook 7H9 broth (Difco) supplemented with  oleic acid, albumin, dextrose and catalase (OADC enrichment  -  BBL/Becton-Dickinson),  to  obtain  final  drug  concentration  ranges of 0.09-25 µg/mL. The isoniazid was dissolved in dis- tilled  water,  and  used  as  standard  drug.  A  suspension  of  the  MTB H37Rv ATCC 27294 was cultured  in Middlebrook 7H9  broth supplemented with OADC and 0.05% Tween 80. The cul- ture was frozen at -80 ºC in aliquots. After two days was carried  out the CFU/mL of a aliquot. The concentration was adjusted  by 5x105 CFU/mL and 100 µL of the inoculum was added to  each well of a 96-well microplate together with 100 µL of the  compounds. Samples were  set up  in  triplicate. The plate was  incubated for 7 days at 37 °C. After 24 h, 30 µL of 0.01% re- sazurin (solubilized in water) was added. The fluorescence of  the wells was read after 24 h in a TECAN Spectrafluor®. The  MIC was defined as the lowest concentration resulting in 90%  inhibition of growth of MTB. Cytotoxicity on mammalian cells In vitro cytotoxicity assays (IC50) were performed first on the  VERO epithelial cells (ATCC CCL81). Compounds with low  cytotoxicity were investigated on the J774A.1 (ATCC TIB-67)  mouse cell line. Both studies as recommended by Pavan et al.  2010  [44].  The  cells  were  routinely  maintained  in  complete  medium (DMEM for VERO and RPMI-1640 (VitroCell®) for  J774A.1) supplemented with 10% heat-inactivated fetal bovine  serum (FBS) plus gentamicin (50 mg/L) and anfotericin B (2  mg/L),  at  37  °C,  in  a  humidified  5%  CO2  atmosphere.  After  reaching confluence, the cells were detached and counted. For  the cytotoxicity assay, 1 × 105 cells/mL were seeded in 200 μL of  complete medium  in 96-well plates  (NUNCtm). The plates  were incubated at same conditions for 24 h, to allow cell adhe- sion  prior  to  drug  testing.  The  compounds  were  dissolved  in  DMSO  (5%)  and  subjected  to  two-fold  serial  dilution  from  1250 to 3.9 μg/mL. Cells were exposed to the compounds at various  concentrations  for  24  h.  Resazurin  solution  was  then  added to  the cell cultures and incubated for 6 h. Cell  respira- tion, as an indicator of cell viability was detected by reduction  of resazurin to resorufin, whose pink colour and fluorescence  indicates cell viability. A persistent blue colour of resazurin is  a sign of cell death. The fluorescence measurements (530 nm  excitation  filter  and  590  nm  emission  filter)  were  performed  in a Tecan Spectrafluor Plus microfluorimeter. The IC50 value  was  defined  as  the  highest  drug  concentration  at  which  50%  of the cells are viable relative to the control. Each test was set  up in triplicate. Selectivity Index The selectivity index (SI) was calculated by dividing IC50 for  the VERO cells by the MIC for the pathogen; if the SI is ≥10, the compound is then investigated further. Acknowledgements The  authors  thank  CYTED  network  (209RT0380),  FAPESP  (process. 2009/06499-1 and 2011/11593-7), ANII for the schol- arship of Mariana Poggi  (BE_INI_2010_1851) and Dr.  Jorge  Castiglioni for the collaboration on the thermogravimetric as- says. References   1. World  Health  Organization,  Global  Tuberculosis  Report  2012,  2012, 3. 2. Koul, A.; Arnoult, E., Lounis, N.;Guillemont, J.; Andries, K. Na- ture 2011, 469, 483-490.   3. Wolff, M.E., Burger’s Medicinal Chemistry and Drug Discovery,  Vol. 2. John Wiley & Sons. Laguna Beach, 1996.   4. Sweetman, S.C., Martindale. Guía completa de consulta farmaco- terapéutica, Pharma Editores. Barcelona, 2008.   5. Gambino, D.; Otero, L. Inorg. Chim. Acta 2012, 393, 103-114.   6. Ruiz-Azuara, L.; Bravo-Gómez, M.E. Curr. Med. Chem. 2010, 17,  3606-3615.   7. Ballin, J.C. JAMA, 1974, 230, 8, 1184-1185.   8. Cuin,  A.;  Massabni,  A.C.;  Leite,  C.Q.;  Sato,  D.N.;  Neves,  A.;  Szpoganicz, B.; Silva, M.S.; Bortoluzzi, A. J. J. Inorg. Biochem. 2007, 101, 291-296.   9. Oliveira,  J. S.; Douza, E. H.S.; Basso, L.A.; Palaci, M.; Dietze,  R.; Santos, D. S.; Moreira, I.S. Chem. Commun. 2004, 312-313.  10. Sousa,  E.H.S.;  Pontes,  D.L.;  Diógenes,  I.C.N.;  Lopes,  L.G.F.;  Oliveira, J.S.; Basso, L.A.; Santos, D. S.; Moreira,  I.S. J. Inorg. Biochem. 2005, 99, 368-375.  11. Oliveira , J. S.; Sousa, E. H. S.; Basso, L. A.; Palaci, M.; Dietze ,  R.; Santos, D. S.; Moreira, Ícaro S. Chem. Commun.2004, 3, 312- 313.  12. Pavan, F. R.; Poelhsitz, G.V.; Barbosa, M. I. F.; Leite, S. R. A.;  Batista,  A.  A.;  Ellena,  J.;  Sato,  L.  S.;  Franzblau,  S.G.;  Moreno,  V.; Gambino, D.; Leite, C. Q. F. Eur. J. Med. Chem.2011, 46, 10,  5099-5107.  13. Maccari, R.; Ottanà, R.;Bottari, B.; Rotondo, E.; Vigorita, M.G.  Bioorg. Med. Chem. Letters, 2004, 14, 23, 5731-5733.  14. Carmo, A.M.L.; Silva, F.M. C.; Machado, P. A.; Fontes, A.P.S.;  Pavan, F. R.; Leite, C.Q.F.; Leite, S. R. De A.; Coimbra, E.S.; Da  Silva, A.D. Biom. Pharm. 2011, 65, 3, 204-209.  15. Tarallo; M.B.; Urquiola, C; Monge Vega, A.; Parajón Costa, B.;  Ribeiro, R. R.;Costa-Filho, A.J.; Mercader, R.C.; Pavan, F.; Leite,  C.Q.; Torre, M.H.; Gambino, D. J. Inorg. Biochem. 2010, 104, 11,  1164-1170.  16. Tarallo; M.B.; Costa-Filho, A.J.;Viera, E.; Monge Vega, A.;Leite,  C.Q.;  Pavan,  F.;  Borthagaray,  G.;  Gambino,  D.;  Torre,  M.H. J. Argent. Chem. Soc. 2009, 97, 1, 80-89.  17. Mondelli, M; Pavan, F.; De Souza; Leite, C.Q.; Ellena, J.; Nasci- mento, O.R.; Facchin, G.; Torre, M.H. J. Mol. Struct.2012, 1036,  180-187.  18. Mondelli, M;Bruné V.; Borthagaray, G.; Ellena,  J.; Nascimento,  O.R.; Leite, C.Q.; Batista, A.A.; Torre, M.H. J. Inorg. Biochem. 2008, 102, 285-292. 204      J. Mex. Chem. Soc. 2013, 57(3)  Mariana Poggi et al. 19. Kremer, E; Facchin, G; Estévez, E.;Alborés P.;Baran, E.J.;Ellena; Torre, M.T. J. Inorg. Biochem. 2006, 100, 1167-1175.  20. Bernstein,  J.;Lott,  W.A.;  Steinberg,  B.S.;  Yale,  H.L.  Am. Rev. Tuberc.1952, 65, 357-364.  21. Middlebrook, G. Am. Rev. Tuberc.1952, 65, 765-767.  22. Geary, W.J. Coord. Chem. Rev.1971, 7, 81-122. 23. Broomhead, J.A.; Kane-Maguire, L.A.P. J. Chem. Soc. A, 1967,  546-553.  24. Lin-Vien,  D.;  Colthup,  N.B.;  Fateley,  W.G.;  Grasselli,  J.G.  The Handbook of Infrared and Raman Characteristic Frequencies of Organic Molecules, Academic Press inc., Boston, 1991. 25. Gunasekaran, S.; Sailatha,E.;Seshadri, S.;Kumaresan, S. Indian J. Pure Ap. Phy. 2009, 47, 12-18. 26. Nakamoto, K. Infrared and Raman Spectra of Inorganic and Co- ordination Compounds. Part  B.  5,  John  Wiley  and  Sons,  New  York, 1997.  27. Tang, X.; Dou, W.; Chen, S.; Dang, F.; Liu, W. Spectrochim. Acta Part A. 2007, 68, 349-353.  28. Nelwamondoa, A. N.;Eveb, D. J.; Watkinsb, G. M.;Brown, M. E.  Thermochim. Acta. 1998, 318, 165-175.  29. Smith, B. Infrared Spectral Interpretation. A Systemic Approach,  CRS Press, Boca Ratón, 1998.  30. Lever, A.B.P., Inorganic Electronic Spectroscopy, Elsevier, New  York, 1984.  31. Cotton, F. A.; Wilkinson, G.; Murillo, C.A.; Bochmenn, M., Ad- vanced Inorganic Chemistry, 6th Ed., Wiley & sons, New York,  1999.  32. Viera,  I.;  Gomez,  M.A.;  Ellena,  J.;  Costa-Filho,  A.J.;Migliaro,  E.R.;  Dominguez,  L.;  Torre,  M.H.  Polyhedron.  2009,  28,  3647- 3653. 33. Facchin, G.; Kremer, E.; Barrios, D.; Etcheverry, S.A.;Costa-Fil- ho, A.J.; Torre, M.H. Polyhedron. 2009, 28, 2329-2334. 34. Das, K.; Sinha, C.; Datta, A.; Garribba, E.; Fondo, M.; Mautner, F.A. ; Fischer, R.C. J. Chem. Res.2012, 36, 12, 722-725.  35. Sanmartin,  J.; Bermejo, M.R.; Garcia-Deibe, A.M.  Inorg. Chim. Acta. 2001, 318, 1-2, 135-142.  36. Urquiola, C.; Gambino, D.; Cabrera, M.; Lavaggi, M.L.;Cerecetto,  H.; Gonzalez, M.; Lopez de Cerain, A.; Monge Vega, A.; Costa- Filho,  A.J.;  Torre,  M.H.  J. Inorg. Biochem. 2008,  102,  1,  119- 126.  37. Bencini,  A.;Gatteschi,  D.,  Electron Paramagnetic Resonance of Exchange Coupled Systems, Springer-Verlag, Berlin, 1990.  38. Pavan,  F.  R.;  Sato,  D.  N.;  Leite,  C.  Q.  F.  An approach to the search for new drugs against tuberculosis. In: Cardona, P.J. (Ed.).  Mycobacterium tuberculosis, book 2. Intech, 2011. 39. Bardou, F.; Raynaud, C.; Ramos, C.; Laneelle, M.A,; Laneelle, G.  Microbiology. 1998, 144, 2539-2544.  40. Scior, T.; Garcés-Eisele, S.J. Curr. Med. Chem. 2006, 13, 2205- 2219.  41. Poole C.F. Thin-layer Chromatography. In: Moffat, A.C.; Ossel- ton, M.D.; Widdop, B.; Watts, J.(Eds.). Clarke’s Analysis of Drugs and Poisons, Pharmaceutical Press, 2011.  42. Leo,  A.;  Hansch,  C.  D.;  Elkins,  J. Chem. Soc. 1971,  71,  525-  616.  43. Palomino,  J.C.;  Martin,  A.,  Camacho,  M.;  Guerra,  H.;  Swings,  J.;  Portaela,  F.  Antimicrob. Agents Chemother.  2002,  46,  2720- 2722.  44. Pavan,  F.R.;  Maia,  P.I.S.;  Leite,  S.R.A.;  Deflon,  V.M.;  Batista,  A.A.;  Sato,  D.N.;  Franzblau,  S.G.;  Leite,  C.Q.F. Eur. J. Med. Chem. 2010, 45, 1898-1905.