Antioxidant cellulose nanofibers/lignin-based aerogels: a potential material for biomedical applications

Lignin is a naturally occurring and aromatic biopolymer with well-known antimicrobial and antioxidant properties. Thus, in this work, the use of cellulose nanofibers (CNF) and lignin to produce ultra-light aerogels for biomedical applications was studied. Aerogels containing varying amounts of lignin (0–30 wt%) and different concentrations of the crosslinking agent Fe3+ (25–100 mM) were developed. The different bioaerogels were fully characterized and their physical, mechanical and bioactive properties analyzed. It was observed that the bioaerogels soluble fraction tends to decrease as the lignin content increases for the different Fe3+ concentrations, due to lignin–CNF interactions through hydrogen bonds. The bioaerogels containing lignin showed remarkable radical scavenging activity as the DPPH concentration decreased with time. This confirms the benefits of including lignin in bioaerogels to impart antioxidant properties. To study the suitability of the produced bioaerogels for controlled drug release, the release of tetracycline (TC) was studied. All of the bioaerogels released TC in a sustained manner for 6 h and presented similar profiles. However, the bioaerogels containing higher concentrations of crosslinker showed a higher release of TC. The TC loading conferred clear antimicrobial activity against S. aureus as expected, unlike the insignificant antimicrobial activity of the bioaerogels without TC. The biocompatibility of the samples was demonstrated for all materials produced (with and without TC loading) by the Kruskal–Wallis test with multiple comparisons. After observation of cell morphology, no significant differences were evident suggesting that the CNF–lignin bioaerogels present optimal biocompatibility for use in the biomedical and pharmaceutical industry. The CNF–lignin bioaerogels presented in this work highlights their promising application as biomedical applications, such as wound dressings due to their biocompatibility, antimicrobial and antioxidant properties, as well as their swelling and solubility properties.


Introduction
Aerogels are porous solid materials that exhibit unique properties making them suitable for application in various fields.Their properties include low bulk density (< 0.5 g cm −3 ), high and interconnected porosity (> 98%, mainly in the mesoporous range), large available specific surface area (> 150 m 2 /g), as well as tunable surface chemistry [1].In recent years, particular attention has been paid to the potential use of aerogels in biomedical applications, including tissue engineering, wound healing, drug delivery, implantable devices, sensing and diagnostics [2,3].However, for the use of aerogels in the development and manufacture of biomedical applications, these materials must demonstrate high biocompatibility, bioactivity, and good biodegradability, in addition to possessing adequate physicochemical properties.
Depending on the nature of the precursors, aerogels are usually divided into three main categories: inorganic, organic and hybrid aerogels [4].Inorganic aerogels are mostly based on metals, oxides and silica, whereas organic aerogels are based on polymers (polystyrenes, polyurethanes, poly alcohols, polyacrylates).In the particular case of employing bio-derived polymers (polysaccharides, proteins, pectins, etc.) the materials are known as bioaerogels [5].Bioaerogels possess cytocompatibility, biocompatibility and biodegradability functionalities, making them significantly more suited to applications in the biomedical sector than inorganic aerogels [6].The biopolymers that constitute the lignocellulosic matrix are outstanding bioaerogel precursors since their processing preserves their bioactivity, biocompatibility and biodegradability, thus facilitating their application in the biomedical sector.Lignocellulosic materials are abundant, renewable and, for the most part, inexpensive resources and materials that are mainly derived from plant biomass, including agricultural waste [7].The structural components of plant biomass are cellulose, hemicellulose and lignin, as well as other types of biopolymers such as proteins, peptides, etc.
Cellulose is the most abundant polysaccharide in nature, whereby it has been used as a replacement material for many oil-based products.In particular, it has attracted increasing interest in the biomedical sector due to its biocompatibility, thermal stability, renewability, biodegradability and harmlessness [8].Two main processes are used to obtain cellulose aerogels: derivatization of cellulose followed by regeneration or by direct dissolution followed by coagulation or precipitation; in both cases forming crystalline forms (allomorphs) of cellulose II [9].Another way to prepare cellulose aerogels without the need for dissolution processes (cellulose I) is the use of cellulose nanostructures (nanocellulose) as a precursor [10].Nanocellulose are cellulose structures that have a dimension of 1-100 nm and can be isolated from natural cellulose fibers.One of the most common types of nanocellulose used in the production of cellulose-based aerogels are cellulose nanofibers (CNF).CNFs are isolated from natural fibers by delamination, i.e., by a strong mechanical treatment, and are 10-20 nm in diameter and several microns in length [11].Owing to their nano-sized diameter and high specific surface area, low concentrations of solid matter are required for the formation of CNF gels (0.5-2%).Considering inherent properties and the nature of nanocellulose, CNF-based aerogels possess several desirable characteristics, including a high specific surface area, mechanical strength, surface activity, porosity, and hydrophilicity.Unlike other applications, parameters such as biocompatibility and non-cytotoxicity are crucial for the development of biomedical applications using CNF-based aerogels.CNF with a high surface charge, such as those obtained through TEMPO-mediated oxidation (TO-CNF), have been found to be noncytotoxic and devoid of irritation potential.Additionally, these CNF exhibit a promising antibacterial effect and can inhibit bacterial swimming properties [12].These properties, combined with their ability to incorporate different materials (such as co-polymers, nanoparticles, etc.) during formulation, make them suitable for various purposes including drug delivery, scaffold fabrication, biosensors, diagnostics, antimicrobial dressings, medical implants, and vascular grafts [10,13,14].
Lignin is the most abundant naturally occurring aromatic macromolecule in nature.It is a natural polyphenolic polymer consisting of three main types of phenylpropane structural units: syringyl propane, guaiacol propane and p-hydroxyphenylpropane units [15,16].It also contains branched active functional groups, such as phenolic, hydroxyl, carboxylic carbonyl and methoxyl groups [17].Lignin's natural origin, much like cellulose, ensures its high abundance, biocompatibility and biodegradability.Thus, presenting itself as an alternative to many synthetic polymers obtained from fossil fuels, for a wide range of applications [18][19][20].In addition to the aforementioned advantages, lignin also exhibits antimicrobial and antioxidant properties [21].These actions occur due to the effects of its phenolic structure on the bacterial envelope causing lysis, as well as its ability to act as hydrogen donor antioxidants [22].All these characteristics make lignin a material with great potential for use in products with biomedical applications such as wound dressing materials [21,23], drug delivery systems [24,25], tissue engineering [26] and pharmaceutical excipients [17,27].The promising use of lignin in the production of aerogels is however associated with certain limitations due to its inability to self-crosslink.Therefore, it is mostly used as an additive mixed with other components such as formaldehyde, alginate, chitosan and poly(ethylene glycol) [25,[28][29][30].
The joint application of CNF and lignin to produce aerogels for biomedical applications is promising due to the desirable properties that these biomaterials exhibit.It has been shown that CNF-lignin hydrogen bonding interaction produces crosslinked 3D-lamellar porous aerogels with increased mechanical performance, density, adsorption and swelling capacity and specific surface area, as well as smaller pore size [31][32][33].This combination of properties paves the way for its use in a multitude of applications in the biomedical and pharmaceutical sector.This study investigates the utilization of cellulose nanofibers (CNF) and lignin to create ultra-light aerogels for biomedical purposes.The physical, mechanical, and bioactive characteristics of various bioaerogels were analyzed by introducing varying quantities of lignin (0-30 wt%) and employing different concentrations of Fe3 + as a crosslinking agent (25-100 mM).Additionally, the release of tetracycline (TC) was evaluated to study the bioaerogels viability for controlled drug release.Furthermore, biocompatibility was established for all bioaerogels created (with or without TC loading) via the Kruskal-Wallis test with multiple comparisons.
The lignin used was a softwood kraft lignin (BioPiva 100) acquired from UPM (Helsinki, Finland).To remove the moisture content of the lignin, it was placed in a forced-air circulated heating oven at 70 °C for 48 h.The characteristics of lignin have been described extensively by the manufacturer and by the authors in previous works [17].The Klason lignin content according to TAPPI T 222 om-02 is about 92% of dry matter and the acid-soluble lignin content according to TAPPI UM 250 is about 4% of dry matter.The sum of both components is commonly considered the value of the total lignin content in the sample (96%).The total amount of carbohydrates according to SCAN-CM 71:09 represents about 2% of the dry matter.The ash content by calcination at 700 °C represents about 1% of the dry matter.The molecular weight of lignin is between 5000 and 6000 Da.

Nanocellulose preparation and characterization
Cellulose nanofibers (TO-CNF) were isolated from bleached wheat straw pulp by TEMPO-mediated oxidation pretreatment.TEMPO-mediated oxidation at 5 mmol/g of cellulose was carried out adding the appropriate amount of 10% NaClO solution as oxidizing agent with continuous stirring at room temperature.A 0.5 M NaOH solution was used to maintain the pH reaction at 10 until no pH decrease was observed (around 2 h).The reaction was stopped by adding 10 mL of ethanol ant the fibers were washed with distilled water until neutral pH.This pretreatment has been proven in previous studies to be the most effective for the production of cellulose nanofibers of nanometric size and high surface charge, favoring their aqueous stability [35,36].After oxidation, a 1 wt% suspension of pretreated fiber was prepared and subjected to a highpressure homogenization treatment in a PANDA 2000 (Gea Niro Soavi, Italy).The nanofibrillation treatment sequence followed was: 4 passes at 300 bar, 3 passes at 600 bar and 3 passes at 900 bar.The TO-CNF obtained were further characterized in accordance to several procedures previously reported by the authors [35][36][37].

Bioaerogels preparation
The bioaerogel samples were crosslinked through the hydrogelation methodology using a trivalent cation [38].Different quantities of lignin (0, 15 and 30 wt% regarding the TO-CNF mass) were added to a TO-CNF aqueous suspension (0.7 wt%), and each mixture was stirred in a paddle mixer for 2 h.The homogenized dispersions (5 g) were then poured into Petri dishes, and 5 mL of Fe +3 solution (25, 50 and 100 mM) was added dropwise carefully along the container's walls.Each Petri dish was then maintained without stirring for 2 h and the resulting hydrogels were soaked and rinsed with a pH 3 HCl solution twice, and then with deionized water three times.Finally, the samples were lyophilized at − 83 °C at a vacuum pressure of 0.5 mBar for 48 h to obtain the nine desired bioaerogel samples.The preparation procedure is illustrated in Fig. 1.The bioaerogels samples were then abbreviated and named according to the amount lignin (0L, 15L and 30L) and the concentration of the Fe +3 solution (25Fe, 50Fe and 100Fe).For instance, the sample 30L-100Fe contained 30 wt% regarding the TO-CNF mass and 5 mL of a 100-mM Fe +3 solution.

Bioaerogels characterization
The different bioaerogels were characterized.Through Fourier transform infrared spectroscopy analysis (FTIR) the possible chemical interactions between the TO-CNF, lignin and Fe 3+ were explored.The FTIR spectra of the prepared materials were collected on an FTIR-ATR Perkin-Elmer Spectrum Two, in the range of 4000-500 cm −1 (employing a resolution of 4 cm −1 and 40 scans for each sample).
The swelling degree (SD %) properties of each sample at t time were determined in distilled water at 25 °C.The obtained data were processed using the following equation (Eq.1): where W i is the weight of the bioaerogel before its immersion, and W t corresponds to the weight of each sample at t time.Measurements were run in triplicate.Additionally, in order to determine the soluble fraction characteristics of the materials, a fixed mass of each bioaerogel was immersed in distilled water at 25 °C for 4 days.Then, each sample was removed from the aqueous medium and further dried until obtaining a constant weight.The soluble fraction (SF %) was calculated as follows (Eq.2): where W i and W f correspond to the weights of the dried bioaerogels before and after immersion, respectively.Measurements were run in triplicate.

Tetracycline release
All the bioaerogels were loaded with TC by placing a disc (8 mm diameter) of the respective bioaerogel into a glass vial containing 2 mL of a TC-deionized water solution (4 mg/mL).The vials were stored in the dark at room temperature for 24 h, to limit any degradation of the TC due to its photosensitivity.Then, discs were removed from the TC solution dried in the dark at room temperature for a duration of ≥ 48 h.The release kinetics of the TC-loaded bioaerogel discs was assessed by placing them into glass vials containing 5 mL PBS, to enable sink conditions.The vials were then incubated at 40 rpm and 37 °C.At defined intervals, 0.5 mL of solution was removed from each vial and the concentration of TC was assessed using a UV-Vis plate reader (PowerWave XS Microplate Spectrophotometer, Bio-Tek, Winooski, USA) at a wavelength of 363 nm, with n = 4 replicates for each condition.After each reading the medium was ( 1) replaced with an equivalent volume of PBS.However, after the reading was taken at 24 h, the entire medium was removed and replaced with fresh PBS.

Antioxidant properties
A DPPH (2,2-diphenyl-1-picrylhydrazyl) radical was employed to measure the antioxidant activity of the different bioaerogel samples based on the radical scavenging property of the lignin [39,40].For this purpose, 5 mL of a DPPH solution dissolved in methanol (50 mg/L) was added to each vial containing a bioaerogel disc (8 mm diameter).A control sample of 50 mg/L of DPPH in methanol was also measured.The glass vials were then stored in the dark at room temperature.At defined time intervals, 300 µL of each sample was collected and the vials were immediately replenished with an equivalent volume of methanol.The absorbance of the different samples was analyzed using a UV-Vis plate reader (PowerWave XS Microplate Spectrophotometer, Bio-Tek, Winooski, VT, USA) at 517 nm, with n = 4 replicates for each condition.The residual DPPH content in the solution was calculated using Eq.3: where A 0 is the absorbance of the control sample and A 1 is the absorbance in the presence of the sample at any time.Decreased absorbance of the reaction indicates a stronger DPPH radical scavenging activity.

Microbiology assay
Bioaerogel discs (8 mm diameter) containing either no TC or loaded with TC solution (4 mg/mL), as previously described, were tested for inhibitory effects on bacterial cultures of Staphylococcus aureus NCTC 10788.For this purpose, Mueller-Hinton (MH) broth was inoculated with S. aureus and then incubated overnight at 37 °C.50 µL of this culture was added to 5 mL of soft MH agar and vortexed to ensure even distribution of the bacteria, the mixture was then poured on top of MH agar plates [41].Bioaerogel discs were then placed on top of the agar and incubated at 37 °C for 24 h.The inhibition zone caused for S. aureus was measured in mm.Furthermore, inoculated plates with S. aureus alone were also incubated as positive control.The results were expressed as mean ± standard deviation, with n = 4 replicates for each condition.

Cytocompatibility test
HEK293T (human embryonic kidney, ATCC) cells were seeded in DMEM (Dulbecco's modified Eagle medium), supplemented with 10% fetal calf serum and (3) non-essential amino acids, in a 24-well plate at 20,000 cells per well and allowed to adhere overnight at 37 °C.An equal volume of culture media pre-incubated with compound was added to each well.Three independent samples of each compound were used.On day 3 posttreatment, cells were imaged on an Olympus widefield microscope using 20 × objective.One representative image is shown from each condition.Cells adhered in all conditions and cell growth was comparable.On day 3 post-culture, cell viability was measured using an MTT assay (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium, Merck).

Statistical analysis
All quantitative data were expressed as mean ± standard deviation.Statistical analysis was performed using a oneway analysis of variance (ANOVA), followed by a multiple comparison test (Tukey's test) to assess whether there were significant differences between the means of the data sets as appropriate.In all cases, p-value < 0.05 indicated a statistically significant difference between tested groups.

Bioaerogel preparation and characterization
By varying both the lignin and the Fe 3+ concentrations, it was possible to successfully obtain nine different bioaerogel samples (Additional file 1: Figure S1).As expected, the different materials did not present any new covalent chemical interactions (Additional file 1: Figure S2).
The bioaerogels swelling properties (Fig. 2) were tested considering both the Fe 3+ and lignin contents.It was expected that the higher the lignin content, the lower the swelling degree capacities due to lignins intrinsic hydrophobicity.The results did not reflect the predicted relationship between lignin and the swelling degree, most likely because the lignin particles were enclosed in the CNF, acting as crosslinking dots.The different Fe +3 concentrations did appear to have some effect on the bioaerogels swelling properties with low lignin contents, however this effect was not easy to spot.The degree of swelling increased as the content of Fe +3 also increased, ranging from 25 to 50 mM.However, higher quantities of cations resulted in the opposite trend.This phenomenon may be explained in terms of the formation of Fe +3 agglomerates, as it was previously observed in other physically crosslinked gel materials [42].Additionally, it is of key importance to keep in mind that this property is strongly dependent on several of the material's network characteristics, such as pore size and morphology, crystallinity degree and gel/soluble fraction [43,44].
The soluble fraction of the different samples appeared to decrease as the lignin content increased for the three considered Fe 3+ concentrations (Fig. 3).These results are in accordance with those expected, since lignin and CNF can successfully interact through hydrogen bonding.Thus, lignin acts as effective crosslinking dots in which CNF are encompassing lignin particles [31,33].Furthermore, the presence or absence of lignin has a stronger effect on this property than the trivalent cation concentrations.The inclusion of Fe +3 only exerts certain undesired negative effect on the materials soluble fraction when considering the highest cation concentration.On the other hand, in the 30L bioaerogel materials the employment of different Fe 3+ concentrations made little difference to the materials solubility.This can be related to the fact that the trivalent cation was added to the CNF/lignin mixture after 2 h time, during which these two components are able to establish a high number of Fig. 2 Swelling degree properties of the bioaerogels Fig. 3 Soluble fraction of the bioaerogels hydrogen bonds, leading to small quantities of CNF capable of further interacting with the Fe 3+ cations.From the obtained results, the three 30L bioaerogel materials were selected to be further tested.

Antioxidant properties of the bioaerogels
In order to assess the antioxidant properties of the ligninbased bioaerogels, a DPPH assay was performed (Fig. 4).All of the lignin-based bioaerogels showed remarkable radical scavenging activity as the DPPH concentration was decreased over time.Thus, these results indicate that the presence of lignin provided all the bioaerogels with antioxidant properties.The 50Fe and 100Fe bioaerogel samples were able to considerably reduce the DPPH concentration up to ca. 84% and 81% after 360 min, and no significant differences were found between these two samples (p = 0.6998).The 25Fe samples exhibited significantly lower antioxidant activity at the same time (74%) in comparison to 50Fe bioaerogel sample (p < 0.05), however, no significant differences were found between 25 and 50Fe bioaerogel samples (p = 0.0897).These results are similar to those found in different studies, where lignin was incorporated within thermoplastic matrices, such as poly(butylene succinate), polylactic acid (PLA) or poly(caprolactone) for achieving antioxidant composite materials [21,39,40].Antioxidant materials, such as these bioaerogels, have the potential to be used for the development of biomedical devices.Materials exhibiting antioxidant properties are able to reduce the concentration of reactive oxygen species (ROS) and free radicals [45,46].The presence of high concentrations of ROS has been shown to interfere and possibly prevent the wound healing process [47,48].Thus, the bioaerogels developed in this study show promising antioxidant properties and therefore could be used as wound dressings to help accelerate the wound healing process.

Tetracycline release from bioaerogels
The release of TC from the bioaerogel discs was evaluated for 24 h.The bioaerogels were capable of providing a sustained TC release for 6 h (Fig. 5).Moreover, it can be seen that all the release profiles had similar shapes.However, bioaerogels containing higher amounts of iron (50Fe and 100Fe bioaerogel samples) showed a higher TC release compared to those containing the lowest amount of iron (25Fe bioaerogel samples).Statistical analysis revealed that there were significant differences between the 25Fe bioaerogel samples and the other 50Fe and 100Fe bioaerogel samples (p < 0.05).However, no significant differences were found between the 50Fe and 100Fe bioaerogel samples (p > 0.05).As previously proposed and reported, TC could bind either Fe 2+ or Fe 3+ [49,50].Therefore, it can be hypothesized that TC could be interacting with iron ions within the bioaerogel matrix.This interaction may result in a higher drug loading for those samples containing a higher amount of iron (50Fe and 100Fe bioaerogel samples) and thus this might explain the difference in drug release observed in this study.The total amount of TC released after 6 h ranged from 115 to 152 µg.This range of released drug was similar to those found in other studies, where different antibioticloaded medical devices were manufactured and exhibited antimicrobial properties [51,52].Therefore, the TC-loaded bioaerogels demonstrated their potential for use as dressings for the treatment of wound infections that if left untreated can significantly delay and disrupt the healing process [53].These outcomes are in line with those discussed in the previous section (antioxidant properties).

Antimicrobial properties of bioaerogels
The antimicrobial activity of TC-loaded bioaerogel discs was tested using a bacterial culture of S. aureus.The results of the disk diffusion experiment are shown in Fig. 6.As expected, all the TC-loaded bioaerogel discs exhibited clear antimicrobial activity against S. aureus (Fig. 6).The diameter of the zone of inhibition in the S. aureus plates containing these TC-loaded bioaerogel discs ranged from 29.88 ± 0.48 mm to 31.13 ± 1.31 mm.However, no significant differences were found between the different bioaerogel formulations used for this purpose (p > 0.05).Moreover, the same bioaerogel formulations containing no TC were also tested on the same bacterial strain to evaluate if the bioaerogels themselves exhibited any antimicrobial properties (Fig. 6A).However, no zone of inhibition was observed in plates containing these control bioaerogel discs without TC.
Although it has been reported that lignin has antimicrobial activity [22,54,55], the presence of lignin was not able to provide sufficient antimicrobial properties to the bioaerogels.The loading of drug molecules, such as TC, into the bioaerogels provided them with notable antimicrobial activity.Once again, the bioaerogels have potential to be used for biomedical applications and, more Fig.6 A picture showing the zone of inhibition obtained for S. aureus in MH agar using bioaerogel discs containing either no TC or loaded with TC solution (4 mg/mL), with n = 4 replicates for each condition Fig. 7 In vitro cell cytocompatibility in HEK293T cells using an MTT assay (A).Cell morphology obtained post-treatment using a 20 × objective (B) specifically, for the manufacture of wound dressings, since these materials could also help prevent infections.

In vitro cytocompatibility of the bioaerogels
In order to ascertain the cytocompatibility of the materials prepared in this work, HEK293T cells were used.This cell line has been used in the past to study the cytocompatibility of different types of biomaterials [56][57][58].Figure 7 shows the MTT test results.The Kruskal-Wallis test with multiple comparisons results show there was no significant difference between the untreated control group and the other samples.Additionally, cell morphology was observed using a widefield microscope 3 days post-treatment with the bioaerogels.In all cases cell morphology was like that observed for the untreated cell group, indicating cell morphology remained unaffected upon treatment with the aerogels.These results suggest that the materials prepared in this work are biocompatible.
These results are not surprising considering that the biocompatibility of the components present in the bioaerogels has been reported in the past.The biocompatibility of TEMPO-oxidized nanocellulose and Fe 3+ has been extensively evaluated suggesting that this material presents excellent cytocompatibility and hemocompatibility [59,60].The biocompatibility of lignin is not as clearly defined, however lignin-based biomaterials have been proven to be cytocompatible or to show low cytotoxicity [61][62][63].Finally, TC has been reported to be biocompatible in several animal studies [64,65].

Conclusions
Studies on the properties of lignin have outlined its potent antioxidant and antimicrobial activity, and its promising application within the biomedical sector; as well as highlighting the limitations caused by its inability to self-crosslink.In this study bioaerogels containing CNF, lignin and the crosslinker Fe 3+ were developed and fully characterized to investigate the bioaerogels properties when different concentrations of lignin and Fe 3+ were introduced.
Lignin exhibited strong radical scavenging activity that gave the bioaerogels potent antioxidant properties capable of reducing ROS concentrations and potentially accelerating wound healing.The bioaerogels swelling degree was unaffected by increasing lignin concentrations, suggesting inclusion of lignin in the CNF masks its intrinsic hydrophobicity.Comparatively, the bioaerogels solubility fraction was decreased by increasing lignin concentration as a result of the lignin-CNF interactions.Despite lignins previously characterized antimicrobial activity, these lignin bioaerogels exhibited no notable inhibition of S. aureus.However, upon loading the bioaerogels with TC, they exhibited potent antimicrobial activity and sustained TC release over 6 h.Inclusion of high Fe 3+ concentrations appeared to promote the bioaerogels release of TC as well as its antioxidant activity, most likely through its crosslinking action.The bioaerogels also demonstrated optimal biocompatibility, highlighting their suitability for biomedical applications.
In conclusion, this study characterizes the antioxidant activity, biocompatibility, TC release, swelling and solubility properties of CNF bioaerogels containing lignin and Fe 3+ , and highlights their promising application as wound dressings.

Fig. 4 Fig. 5
Fig. 4 Residual DPPH content for the lignin-based bioaerogels using an initial DPPH concentration of 50 mg/L at 360 min, with n = 4 replicates for each condition