Abstract
Plastic products have become indispensable in various industrial sectors and everyday life worldwide. However, conventional plastics present significant sustainability challenges due to their reliance on unsustainable petroleum sources, contribution to greenhouse gas emissions, and their resistance to biodegradation, which leads to environmental waste accumulation. In response to these issues and the need to decrease carbon emissions, there has been a growing interest in bioplastics. One promising family of bioplastics is polyhydroxyalkanoate (PHA), a group of natural polyesters that can serve as a sustainable alternative to traditional plastics in numerous applications. PHA stands out among bioplastics for several reasons, including production from organic waste substrates. The polyester can be produced from various organic waste materials, reducing the demand for costly virgin resources and providing an environmentally friendly solution for waste management. PHA additionally possesses high biodegradability compared to other bioplastics on the market, allowing it to break down more efficiently in the environment and minimize long-term pollution. PHA can replace several families of petroleum-based plastics, making it a versatile option for a wide range of products and applications. However, despite these promising attributes, the high production cost of PHA remains a significant barrier to its widespread adoption. The expensive carbon sources utilized in microbial fermentation are the primary factor contributing to this cost, limiting the growth of the PHA market. There is a critical need to develop an efficient bioprocessing system that can produce high-quality PHA at a significantly lower cost. To address this imperative, the primary goal of this study was to create an integrated PHA production system that utilizes inexpensive cheese byproducts feedstock, making PHA production more economically viable while maximizing resource efficiency and minimizing waste.
Whey and whey byproducts generated from cheese production offer a promising source for bioplastic production. Researchers estimate that approximately half of this whey produced from cheese making is processed into useful forms for human or animal consumption, with the rest ending up as waste material, contributing to high carbon emissions. While some cheese manufacturers can convert whey byproducts into valuable food ingredients, like whey protein concentrates and lactose powder, additional byproducts continue to be generated. Whey permeate is produced after concentrating the liquid whey with membrane filtration, containing most of the lactose and minerals. The whey permeate can be further processed into lactose powder with evaporation and drying systems creating byproduct delactosed permeate (DLP). Crystallization limitations prevent the complete recovery of lactose for food products. As a result, approximately one-third of the lactose remains as delactosed permeate (DLP), which is either sold as low-value animal feed or disposed of as waste. DLP is the main cheese byproduct without a widespread commercial application, which presents an opportunity for biomaterial production. Therefore, this study aimed to convert low-value cheese byproducts DLP into high-value bioplastic PHA.
The PHA producer Haloferax mediterranei (H. mediterranei) appears to be a suitable candidate for utilizing cheese byproducts as feedstock. This fit is due to the archaea’s unique ability to thrive in extreme halophilic conditions, tolerating salinity levels of up to 20%. The presence of high salt-containing cheese byproducts provides a compatible source of nutrients for H. mediterranei, compared to other PHA producers and common contaminant organisms that can become inhibited from such high salt concentrations. The high salt environment removes the need for an expensive sterility unit operation, such as high-temperature exposure or antibiotics, reducing production costs. Moreover, the intracellular PHA extraction from H. mediterranei can be achieved through a straightforward process of water addition, causing osmotic shock and cell lysis. High salt-containing cheese byproducts provide Another advantage of H. mediterranei is its ability to produce a higher-value PHA polyester, poly(3-hydroxybutyrate-co-3-hydroxyvalerate) (PHBV). This PHA polyester exhibits enhanced properties compared to the more commonly produced polyhydroxybutyrate (PHB). The presence of 3-hydroxyvalerate units in PHBV enhances its flexibility, making it suitable for various applications, especially in the food industry. The potential applications of PHBV in food packaging, such as films, bags, containers, or cutlery, align with the circular economy principles. By utilizing cheese byproducts and producing high-value PHBV, the cheese industry can take significant steps toward sustainability and waste reduction.
A pretreatment method of the DLP was developed and optimized utilizing centrifugation and enzymatic lactase hydrolysis. The centrifugation pretreatment was shown to increase lactase hydrolysis efficiency and H. mediterranei cell growth. The optimal lactase enzyme conditions with the centrifuged DLP to achieve a lactose hydrolysis efficiency of 92.5% were determined to be a loading of 0.025 g lactase/g lactose and shaking incubation at 40 ℃ for 12 hours. The effects of hydrolyzed DLP on H. mediterranei cell growth and PHA production were studied. The hydrolyzed DLP achieved high PHA yields of 0.33 ± 0.01 g VS/g total sugar that surpassed yields obtained with glucose feedstock. However, lower galactose consumption by H. mediterranei was observed to limit PHA production with this substrate. After demonstrating that hydrolyzed DLP could result in effective H. mediterranei growth and PHA production, nitrogen source optimization was conducted to devise a cost-effective feedstock for H. mediterranei growth on hydrolyzed DLP substrate. Three nitrogen sources were tested at equivalent C/N ratios of 8, yeast extract, ammonium chloride, and an equal nitrogen loading of yeast extract and ammonium chloride. Supplementation of ammonium chloride resulted in the highest PHA yield, reaching 0.34 ± 0.03 g VS/g total sugar. Furthermore, the media supplemented with ammonium chloride nitrogen source gave the lowest PHA production of the feedstocks investigated. These cost savings make ammonium chloride an economically favorable choice for PHA production compared to other nitrogen sources containing yeast extract. The results of these studies demonstrated that cheese byproduct DLP could be used as an effective feedstock for PHA production, helping to create a circular economy for the cheese industry.
The hydrolyzed DLP supplemented with ammonium chloride nitrogen source was studied as feedstock for H. mediterranei cultivation with three feeding strategies; batch, fed-batch, and continuous feeding. With batch feeding, four different hydrolyzed DLP and whey permeate loadings of 10, 20, 30, and 40 g total sugar/l were studied. The loading of 10 g total sugar/l was determined to be the maximum loading of hydrolyzed DLP, as higher loadings resulted in considerable inhibition. However, hydrolyzed whey permeate resulted in a higher maximum loading of 30 g total sugar/l. This difference in maximum loading between the two hydrolyzed cheese byproduct feedstocks indicated that inhibitory constituents were present in DLP that were not in whey permeate. High-heat reaction products that form during whey permeate processing to lactose powder and DLP, such as Maillard products that are inhibitory to other microorganisms, could be the source of increased inhibition with DLP. The PHA yields of the hydrolyzed DLP and whey permeate were similar at 0.30 ± 0.03 and 0.27 ± 0.02 g VS/g total sugar, respectively. These yields were on the higher end of those reported for glucose feedstock of 0.07-0.33 g PHA/g sugar, demonstrating that hydrolyzed cheese byproducts are an effective substrate for PHA production. The maximum specific growth rates of H. mediterranei growth with hydrolyzed DLP and whey permeate were estimated with batch feeding and found to be similar for the two byproducts. These results can assist in developing large-scale PHA production facilities utilizing the cheese byproduct substrates.
Based on the batch operation findings, a 6 L fed-batch bioreactor with an initial working volume of 3 L was operated to attempt to decrease substrate inhibition with the hydrolyzed DLP. Hydrolyzed DLP and whey permeate feedstocks were fed every four days at a loading of 10 g total sugar/l. A 50% increase in the final PHA concentration of the hydrolyzed DLP was achieved with the fed-batch operation and two feedings than with the single batch loading of 10 g total sugar/l. This increase in PHA production with fed-batch operation was due to the additional loading of 10 g total sugar/l that was able to be fed without inhibition occurring. However, the culture became inhibited and unstable past the third feeding of hydrolyzed DLP. For the whey permeate, the PHA concentration increased with the first three feedings, and a stable culture was maintained after three additional feedings. These results confirmed that increased inhibition occurs with hydrolyzed DLP compared to hydrolyzed whey permeate feedstock. The results indicated that fed-batch bioreactor operation with hydrolyzed DLP and hydrolyzed whey permeate feedstocks can improve PHA production compared to batch cultivation.
Since fed-batch feeding only increased PHA production with hydrolyzed DLP marginally, a continuous stirred tank reactor (CSTR) with hydrolyzed DLP feedstock was studied to try to reduce substrate inhibition further and increase PHA production. The CSTR was operated at a hydraulic retention time of 10 days with an organic loading rate (OLR) of 2.5 g total sugar/l/day, and a hydraulic retention time of 20 days with various OLRs of 1.25-2 g total sugar/l/day. The 10-day HRT CSTR resulted in a PHA yield of 0.18 g VS/g total sugar, comparable to that achieved with the fed-batch bioreactors. However, the 10-day CSTR became unstable after 22 days of continuous feeding, potentially due to hydrolyzed DLP substrate inhibition. Although the 20-day HRT CSTR was able to maintain a stable culture for 130 days, considerably lower PHA yields were achieved of < 0.10 g VS/g total sugar. Therefore, continuous feeding at the applied HRTs and OLRs was not recommended for PHA production with hydrolyzed DLP.
Lower galactose consumption was observed for the batch and fed-batch studies than for glucose. Therefore, a galactose acclimation study was conducted to determine if growing H. mediterranei on successive batches of galactose could increase consumption and PHA production with the sugar. The archaea was cultivated on galactose for five consecutive batches. The final cell dry mass (CDM) concentration of H. mediterranei significantly increased with three batches, the specific growth rate with four batches, the final PHA concentration with five batches, and the galactose consumption with three batches (total galactose consumption), demonstrating effective galactose acclimation had occurred. The research can potentially lead to improved PHA yields with galactose-containing feedstocks such as hydrolyzed cheese byproducts. Overall, the study demonstrated that cheese byproduct delactosed permeate could successfully be utilized as feedstock for PHA bioplastic production with H. mediterranei. The results of the study should be used to further scale the proposed process leading to commercialization.