兽研引言
宠物人性化趋势日益增强，人们的关注点已转向为伴侣动物提供最佳营养，这些动物如今常被视为家庭成员[1]。这一转变伴随着人们对宠物食品中传统蛋白质来源的可获得性及其环境影响的担忧日益增加。由于人口增长，预计到2050年全球肉类需求将增加75%[2]。值得注意的是，这一预测并未考虑全球宠物数量的同步增长，而宠物数量的增长进一步放大了对富含蛋白质日粮的需求，以满足动物的营养需要。必需氨基酸摄入不足可能导致营养缺乏及相关健康并发症[3]。
随着全球蛋白质需求持续上升，而传统来源的产量停滞不前，具有较低环境足迹的替代蛋白质作为可行解决方案受到了关注[4]。微生物蛋白已成为人类和动物营养中传统动物蛋白的有前景替代品[5]。这些蛋白质通过利用真菌、酵母和细菌等微生物的高效且可扩展的发酵工艺生产，具有营养和环境效益的独特组合。先前的研究表明，微生物蛋白具有高消化率和优异的氨基酸谱，使其在营养上与传统蛋白质来源相当甚至更优[6,7]。这些特性对于满足伴侣动物的膳食需求尤为有利，因为伴侣动物依赖均衡的氨基酸摄入以维持最佳健康。此外，与传统蛋白质来源相比，微生物蛋白的生产所需农业用地和水显著更少，同时温室气体排放也更少，使其成为应对可持续蛋白质生产挑战的有吸引力的解决方案[5]。
然而，在宠物食品领域，此类替代品的成功不仅取决于可持续性和可获得性。可行的蛋白质来源必须满足营养标准，同时还要对宠物具有适口性[4]。这种环境、营养和感官标准的整合对于确保替代蛋白质在宠物食品配方中的市场成功至关重要。替代蛋白质的接受度在很大程度上受味觉、质地和香气等感官特征的影响[4]。因此，适口性是确保宠物乐于接受这些新型蛋白质来源的关键决定因素。因此，适口性试验对于评估宠物在其日粮中如何感知这些蛋白质，以及促进其无缝融入宠物食品而不损害感官吸引力或整体适口性而言不可或缺[8]。评估宠物食品的适口性通常涉及在犬舍等受控环境中或通过家庭使用测试（IHUT）在自然家庭环境中进行的标准化“消费”测试。这些测试在所用动物的品种和体型方面具有灵活性，使研究人员能够收集跨不同宠物群体的数据[4,8]。常见方法包括单碗接受度测试和双盘偏好测试，用于评估食物的吸引力和偏好。接受度定义为动物从某一食品中摄入足够热量以维持健康和表现的程度，与味觉或嗅觉等感官特征无关[4,8]。相比之下，偏好反映的是对一种食物相对于另一种食物的比较性选择，受香气和质地等感官因素的影响[9]。香气的作用尤为显著，因为犬和猫高度依赖其嗅觉，并受犁鼻器（雅各布森器官）增强，该器官可放大气味感知，并受湿度和温度等环境因素的影响[4]。“首选”反应以及动物对食物的初始接近和采食常用于评估香气的影响。有时，动物会因先前负面经历导致的“条件性厌恶”而拒绝其他动物已接受的食物[10]。行为反应如“新食物恐惧症”（对新食物的恐惧）和“嗜新症”（对新食物的偏好）经常被观察到，尤其在猫中，但在犬中也是如此[4]。开发宠物食品替代蛋白质来源需要平衡可持续性、营养和适口性。解决这些因素对于减少传统肉类生产的环境影响，同时确保宠物食品对伴侣动物仍然具有吸引力和营养性至关重要[3,4,8]。
本研究旨在通过犬猫双碗家庭使用测试（IHUT）评估含微生物蛋白的干性宠物食品的短期偏好性。本研究探讨以微生物蛋白替代传统动物蛋白源对犬猫干性膨化日粮适口性的影响。通过应用成熟的适口性测试方法并评估动物对新食物的行为反应，我们旨在确定这种替代是降低、维持还是提高食物偏好性。此外，对实验日粮进行了独立实验室特性分析，以支持在营养可比配方框架内对观察到的适口性结果进行解读。本研究建立在先前关于微生物蛋白高消化率和平衡氨基酸组成的发现基础上，同时强调其对更可持续宠物营养的潜在贡献。我们的核心假设是，微生物蛋白可以纳入宠物食品配方而不影响适口性，甚至可能提高犬猫的食物偏好性。
材料与方法
研究设计
适口性试验采用家庭使用测试（IHUT），以获得更具代表性的宠物偏好性结果[11]。本研究按照独立合同研究组织（Sense Test，葡萄牙）常规使用的标准化IHUT程序设计，反映了伴侣动物适口性评估的既定行业实践。每种动物纳入30只，与商业家庭使用测试（IHUT）适口性研究中常用的样本量一致。所有动物均登记在包含品种和体重等详细特征的数据库中。犬年龄3至16岁，体重3至10 kg（平均6 kg），涵盖11个品种。猫年龄3至17岁，体重2至10 kg（平均5 kg），涵盖6个品种。
所有动物均已熟悉受试食物类型，即商业可获得的工业化生产干性食品，这与IHUT条件下开展的适口性试验中通常采用的做法一致[11]。根据主人报告，所有动物均健康，无慢性疾病或食物过敏史。
试验期间，每只动物接受与其日常习惯采食量相同的食物量。饲喂量根据年龄和体重进行调整，以确保与正常饮食摄入模式一致。该方法旨在最大限度减少与采食习惯相关的变异性，同时维持动物的营养需求，符合标准IHUT方法学[11]。
对照与试验配方
适口性试验评估了相同的两种干性膨化日粮：对照日粮和含微生物蛋白的试验日粮。试验日粮以10%（w/w）水平添加微生物蛋白配制，以下称为含MP日粮。对照日粮和含MP日粮的配制旨在实现可比的营养目标，主要区别在于蛋白质来源。生产后观察到的小的分析差异反映了正常的配方和分析变异性。
对照日粮的配制模仿了以鱼和米饭风味特征销售的商业可获得全价均衡配方。其组成包括谷物（12%大米）、肉类和动物衍生物（34%）、禽脂、磷虾（3%）、甜菜粕（1.6%）、水解动物蛋白、酵母、角豆、矿物质、维生素和丝兰（0.001%）。
含MP日粮在总体营养组成上与对照日粮相同，但在原料组成上有所不同，即添加了MP，完全替代鱼粉并部分替代禽肉粉。由于微生物蛋白本身脂肪含量低，对禽脂进行了调整以维持日粮之间可比的营养目标。两种日粮的组成见表1，生产后测定的分析成分见表2。
Table 1. Ingredient composition of experimental pet food formulations used in palatability assessment.
| Ingredient composition (%) | ||
|---|---|---|
| Control Formula (Fish and Rice) | Test Formula (containing MP) | |
| Cereals | 51% | 51% |
| Poultry Meal | 28% | 22% |
| MP | – | 10% |
| Fish Meal | 3.5% | – |
| Meat and bone Meal | 2% | 2% |
| Poultry fat | 7% | 7.5% |
| Krill | 3% | 3% |
| Beet pulp | 2% | 2% |
| Hydrolyzed animal protein | 2% | 2% |
| Carob | 0.2% | 0.2% |
| Yeast | 0.3% | 0.3% |
| Premix | 0.5% | 0.5% |
| Yucca schidigera | 0.001% | 0.001% |
Table 2. Analytical composition of experimental diets and estimated metabolizable energy.
| Analytical Composition | ||
|---|---|---|
| Control Formula (Fish and Rice) | MP-containing diet | |
| Crude Protein | 30% | 30% |
| Crude Fat | 13% | 13% |
| Crude Fiber | 2.5% | 2.5% |
| Raw Ash | 8.5% | 9% |
| Moisture | 8% | 8% |
| Estimated metabolized energy (kcal/kg) | 3,485 | 3,468 |
两种实验日粮均在Ovargado S.A.（葡萄牙圣若昂德奥瓦尔）采用工业规模单螺杆挤压系统（Andritz）在受控且相同的条件下生产，最小批次产能约为3，000 kg。基础混合物通过直接蒸汽注入和加水进行预处理，以促进淀粉糊化和初步热处理。挤压在受控的热学和机械参数下进行，机筒温度为120 °C至140 °C。挤出物在模头处切割形成均匀的颗粒，随后干燥至最终水分含量低于10%。
挤压后，干燥的颗粒使用旋转滚筒包衣机在环境温度和大气压力下进行包衣。首先施加部分禽脂（占配方的7.0–7.5%），然后施加粉末状禽类消化物（0.2%，w/w）作为适口性增强剂。对照组和含MP日粮均采用相同的挤压后包衣程序，包括相同的禽脂和禽类消化物施加量。
测试地点和产品配送
实验试验在动物的习惯性饲喂环境中进行，主要在其家庭环境（葡萄牙家庭）中。测试产品由Sense Test， Lda.（葡萄牙加亚新城）的技术人员或代表递送给参与者。在个别会面中，解释了试验条件、方案和要求，并向参与者提供了书面说明，参与者须在开始试验前确认。宠物主人被指示将日粮作为宠物日常饲喂的一部分进行喂养，以确保一致性。产品装在适当的容器中递送，每个容器清楚标有代码和样品的通用描述。所有样品均进行编码以保持盲法并在整个试验过程中尽量减少偏差。使用编码标签确保参与者在研究期间不了解日粮的具体配方，从而减轻对饲喂行为和试验结果的任何潜在主观影响。
双碗偏好测试
双碗偏好测试用于评估短期食物偏好，操作如下：每只动物同时呈现两碗食物，一碗含MP日粮，另一碗含对照日粮[4，8]。测试连续两天进行（每只动物四餐），代表食物偏好的短期评估。为尽量减少潜在偏差，在四餐中每次交替碗的位置。两个碗均放置在动物的进食区域，两份样品的食物分量标准化为相同数量，对应于动物日常餐食中通常提供的量。主人被指示在两天研究期间不提供零食、餐桌残渣或任何其他食物，并使用Sense Test提供的食碗。试验在常规家庭饲喂条件下进行，以尽量减少对动物通常进食行为的干扰。
数据收集和分析
每餐时，主人使用数字秤记录空碗、饲喂前含食物的碗以及饲喂后含剩余食物的碗的重量。食物消耗量计算为每餐提供食物量与剩余食物量之间的差值。每只动物的四餐食物消耗数据在统计分析前进行汇总，每只动物被视为实验单位。主人还记录了首先接近的碗、动物首先从中进食的碗，以及测试期间总食物摄入量是否高于、低于或类似于动物的通常摄入量。这些观察结果进行描述性总结。统计分析使用XLSTAT（Lumivero， USA）和Microsoft Excel进行。含MP日粮与对照日粮之间食物摄入量的差异使用Wilcoxon符号秩检验评估。统计显著性设定为p < 0.05。首次接近和首次进食以百分比进行描述性报告。
实验日粮的分析表征
进行了独立实验室分析以表征对照日粮和含MP日粮，并支持对适口性结果的解释。所有分析均由Eurofins Food Testing Lisboa（葡萄牙里斯本）使用经验证的内部方法或国际公认的标准方法进行，具体如分析报告中所述。
近似成分，包括粗脂肪（酸水解后）和粗灰分，使用内部重量法测定。脂肪酸谱通过气相色谱-火焰离子化检测（GC-FID）分析，结果以相对百分比和基于脂肪含量的绝对值（g/100 g）表示。
氨基酸组成在酸水解后根据ISO 13903:2005使用离子色谱-紫外检测（IC-UV）测定。含硫氨基酸（半胱氨酸和胱氨酸）和蛋氨酸在氧化水解后分析，而色氨酸根据EU Regulation 152/2009使用液相色谱-荧光检测（LC-FLD）单独定量。
采用经内部验证的方法（CON-PV 00006，2023−12），通过电感耦合等离子体发射光谱法（ICP-OES）对选定矿物质进行定量分析。
为支持对适口性结果的解读，依据内部实验室方法，采用气相色谱-质谱联用技术（GC-MS）对香气相关挥发性化合物进行了针对性筛查。
结果
犬粮
采用30只不同品种犬（平均体重5.9 ± 2.4 kg）开展了一项IHUT试验，以评估含或不含微生物蛋白配方的短期食物偏好性。两种日粮的配方均达到可比的营养目标，主要差异在于蛋白质来源。总体食物摄入量与动物平时的摄入量相近，91% ± 3%的参与者报告其犬采食量达到或超过平时的常规餐量。仅4名参与者（13%）报告粪便稠度出现轻微改变，其中2名报告粪便较硬，2名报告粪便较软。
在食物偏好性方面，双碗偏好试验显示，与对照日粮相比，犬对含MP日粮表现出统计学显著的偏好。在四餐中，58%的犬首先接近含MP日粮，58%的犬首先采食该日粮，而对照日粮分别为42%（图1和图2）。图3显示，犬采食含MP日粮的比例显著高于对照日粮（72% vs. 50%；Wilcoxon符号秩检验，p = 0.002）。
[图：首次接近（犬）。犬首次接近含MP日粮或对照日粮的观察百分比]
[图：首次采食（犬）。犬首次采食含MP日粮或对照日粮的观察百分比]
[图：食物摄入量（犬）。双碗偏好试验期间犬对含MP日粮和对照日粮的平均消费比例]
猫粮
开展了类似的IHUT试验，但此次使用30只猫（平均体重5.3 ± 2.1 kg）。据报告，四分之三的猫（75% ± 2%）在试验期间维持或增加了食物摄入量，高于其平时的常规餐量。总体而言，观察到两种日粮之间的食物偏好性存在统计学显著差异（Wilcoxon符号秩检验，p = 0.001）。
平均而言，在提供的四餐中，62%的猫首先朝向盛有含MP日粮的碗（图4），68%的猫首先采食该日粮（图5）。含MP日粮的平均摄入量达到所提供食物总量的51%，而对照日粮为34%（Wilcoxon符号秩检验，p = 0.001；图6）。
[图：首次接近（猫）。猫首次接近含MP日粮或对照日粮的观察百分比]
[图：首次采食（猫）。猫首次采食含MP日粮或对照日粮的观察百分比]
[图：食物摄入量（猫）。双碗偏好试验期间猫对含MP日粮和对照日粮的平均消费比例]
对照日粮和含MP日粮的分析表征
独立实验室分析显示，含微生物蛋白日粮和对照日粮的配方旨在达到可比的营养目标，而非完全相同的分析组成，如表3所示。观察到轻微的分析差异，包括灰分、色氨酸和ω6:ω3比值，总体上维持了广泛可比的营养特征。粗脂肪分析值相当（两种日粮均为13.1%），而含微生物蛋白日粮的灰分含量略高（9.8% vs. 对照的9.0%）。两种日粮的氨基酸谱总体可比。谷氨酸水平在两种日粮间相当（3.73 vs. 3.71 g/100 g），而含微生物蛋白日粮的色氨酸含量较高（0.333 vs. 0.259 g/100 g）。脂肪酸谱总体相似；然而，含微生物蛋白日粮的omega-6:omega-3比值较低（8.3 vs. 对照的11.2）。针对性挥发性化合物筛查未发现任一日起粮中单个挥发性化合物出现显著升高，大多数分析物检测值低于定量限。
Table 3. Selected analytical characteristics of the control and MP-containing diets.
| Parameter | Control Diet | MP-Containing Diet |
|---|---|---|
| Crude Fat (%) | 13.1 | 13.1 |
| Ash (%) | 9.0 | 9.8 |
| Glutamic Acid (G/100 G) | 3.71 | 3.73 |
| Tryptophan (G/100 G) | 0.259 | 0.333 |
| Ω 6: Ω 3 Ratio | 11.2 | 8.3 |
| Targeted Volatile Screen | Mostly <Loq | Mostly <Loq |
讨论
本研究采用双碗偏好测试，在家庭使用测试（IHUT）方案下评估了含MP的干性膨化日粮的短期食物偏好，并与营养等效、宏量营养素匹配的对照日粮进行比较，该对照日粮的配方主要差异在于蛋白质来源。双碗测试是宠物食品适口性研究中广泛建立的方法学，因其能够提供可靠且可重复的膳食偏好测量而受到认可[4]。动物可能因习惯、空间舒适度或其他环境因素而对采食区域的特定一侧形成偏好，这可能使适口性测试结果产生偏倚。通过交替放置食碗，研究者将位置偏倚的影响降至最低，有助于将观察到的偏好解释为与食物本身的感觉属性（如味道、香气或质地）相关，而非碗位置等外部因素[4]。大多数研究参与者报告的总采食量与宠物（猫和狗）通常摄入量相似甚至更高。研究指出宠物主人能够准确识别和解读其动物的采食行为，且这与通过定量食物消耗测得的适口性水平强相关[12]，这强调了主人报告观察结果在这些试验中的重要性。实验设计的一个重要方面是微生物蛋白完全替代了鱼粉，鱼粉通常被认为是伴侣动物日粮中适口性很高的原料。尽管进行了这一替代，犬和猫在所评估的条件下均对含MP日粮表现出显著偏好。这些发现表明，在所评估的条件下，用微生物蛋白替代鱼粉并未损害短期食物偏好。结果表明，研究方案未实质性改变动物的日常采食规律，支持了在常规家庭喂养条件下对观察到的食物偏好的解释。选择四餐评估期旨在减少初始新奇反应的可能影响，同时保持偏好评估的短期性质。尽管重复暴露可能减弱短暂的新食性或恐新性反应，但本研究并非为独立表征这些行为现象而设计，因此观察到的偏好应在短期食物偏好测试的背景下进行解释。本研究的发现表明，在所评估的条件下，含MP的日粮优于对照日粮。这些发现支持进一步研究微生物蛋白作为宠物食品替代蛋白质来源的潜力。
结论
总之，本研究表明，添加10%微生物蛋白未损害短期食物偏好，且在所评估的条件下与偏好增加相关。与对照日粮相比，犬和猫均偏好含MP的日粮，这提供了短期食物偏好的新证据，补充了先前报道的营养质量和消化率[6,7]。纳入犬和猫的数据是一项重要贡献，因为替代蛋白质来源的适口性研究历来主要集中于犬。这些发现突显了MP作为适用于多种伴侣动物物种的多功能原料的潜力。综合来看，这些发现支持进一步研究MP作为宠物食品配方中的替代蛋白质来源，同时结合其先前报道的营养质量、受控生产系统的稳定供应[6]以及良好的食物偏好。由于本研究仅评估了短期食物偏好，无法就长期喂养行为、持续接受度或健康结局得出结论。需要进行涉及不同生命阶段和品种的长期喂养试验的进一步研究，以确认这些发现并扩展其在伴侣动物营养中的适用性。
支持信息
[内容请查看原文]
[内容请查看原文]
[内容请查看原文]
[内容请查看原文]
查看英文原文▾
Introduction
The increasing trend of pet humanization has shifted the focus towards providing optimal nutrition to companion animals, which are now often regarded as family members [1]. This shift is accompanied by rising concerns about the availability and environmental impact of traditional protein sources used in pet food. Global meat demand is projected to increase by 75% by 2050 due to population growth [2]. Notably, this projection does not account for the concurrent increase in the global pet population, which further amplifies the demand for protein-rich diets to meet animals’ nutritional requirements. Insufficient intake of essential amino acids may lead to nutritional deficiencies and associated health complications [3].
As global protein demand continues to rise amidst stagnant production volumes of traditional sources, alternative proteins with lower environmental footprints have gained attention as viable solutions [4]. Microbial proteins have emerged as promising alternatives to traditional animal proteins in both human and animal nutrition [5]. Produced through efficient and scalable fermentation processes using microorganisms like fungi, yeast, and bacteria, these proteins offer a unique combination of nutritional and environmental benefits. Prior studies have demonstrated that Microbial proteins exhibit high digestibility and an excellent amino acid profile, making them nutritionally comparable or even superior to conventional protein sources [6,7]. These characteristics are particularly advantageous for meeting the dietary requirements of companion animals, which depend on balanced amino acid intake for optimal health. Additionally, Microbial proteins production requires significantly less agricultural land and water while emitting fewer greenhouse gases compared to conventional protein sources, making them an attractive solution for addressing the challenges of sustainable protein production [5].
In the context of pet food, however, the success of such alternatives depends on more than just sustainability and availability. A viable protein source must meet nutritional standards while also being palatable to pets [4]. This integration of environmental, nutritional, and sensory criteria is critical for ensuring the market success of alternative proteins in pet food formulations. The acceptance of alternative proteins is heavily influenced by sensory characteristics such as taste, texture, and aroma [4]. As a result, palatability is a critical determinant in ensuring that pets readily accept these novel protein sources. Consequently, palatability trials are indispensable for evaluating how pets perceive these proteins within their diets and for facilitating their seamless incorporation into pet foods without compromising sensory appeal or overall palatability [8]. Evaluating the palatability of pet foods typically involves standardized “consumption” tests conducted in controlled environments like kennels or in a natural home setting via in-home use tests (IHUT). These tests offer flexibility in terms of the breeds and sizes of animals used, allowing researchers to gather data across diverse pet populations [4,8]. Common methods include single-bowl acceptance tests and two-pan preference tests, which assess food appeal and preferences. Acceptability is defined as the extent to which an animal consumes sufficient calories from a food item to maintain health and performance, independent of sensory characteristics such as taste or smell [4,8]. Preference, by contrast, reflects a comparative choice for one food over another, influenced by sensory factors like aroma and texture [9]. Aroma plays a particularly significant role, as dogs and cats rely heavily on their olfactory senses, enhanced by the vomeronasal organ (Jacobson’ s organ), which amplifies scent perception and is influenced by environmental factors like moisture and temperature [4] The “first-choice” response and animal’s initial approach to and consumption of a food is often used to assess the impact of aroma. Occasionally, animals reject foods accepted by others due to “conditioned aversion” resulting from prior negative experiences [10]. Behavioral responses like “neophobia” (fear of new foods) and “neophilia” (preference for novel foods) are frequently observed, particularly in cats but also in dogs [4]. Developing alternative protein sources for pet food requires balancing sustainability, nutrition, and palatability. Addressing these factors is essential to reduce the environmental impact of traditional meat production while ensuring that pet food remains both desirable and nutritious for companion animals [3,4,8].
The objective of this study was to evaluate the short-term preference for dry pet food containing microbial protein using a two-bowl in-home use test (IHUT) in dogs and cats. This study investigates the impact of replacing traditional animal protein sources with MP on the palatability of dry extruded diets for dogs and cats. By applying established palatability testing methods and evaluating behavioral responses to novel foods, we aim to determine whether this substitution decreases, maintains, or enhances the food preference. In addition, independent laboratory characterization of experimental diets was conducted to support interpretation of the observed palatability outcomes within a nutritionally comparable formulation framework. This work builds on previous findings regarding the high digestibility and balanced amino acid profile of microbial proteins, while emphasizing their potential contribution to more sustainable pet nutrition. Our central hypothesis is that microbial protein can be incorporated into pet food formulations without compromising palatability and may even increase the food preference in dogs and cats.
Materials and Methods
Study design
For the palatability trial, In-Home Use Testing (IHUT) was used to allow for more representative outcomes of pet preferences [11]. The study was designed according to the standardized IHUT procedures routinely used by the independent contract research organization (Sense Test, Portugal), reflecting established industry practice for companion animal palatability assessments. Thirty animals per species were enrolled, consistent with the sample size commonly used in commercial in-home use test (IHUT) palatability studies. All animals were registered in a database that included detailed characteristics such as breed and body weight. Dogs ranged from 3 to 16 years of age, had a body weight between 3 and 10 kg (mean 6 kg), and represented 11 breeds. Cats ranged from 3 to 17 years of age, weighed between 2 and 10 kg (mean 5 kg) and represented 6 breeds.
All animals were familiarized with the type of food being tested, specifically commercially available industrially produced dry food, as commonly applied in palatability trials conducted under IHUT conditions [11]. According to owner reports, all animals were healthy and had no history of chronic diseases or food allergies.
During the trial, each animal received the same amount of food they were accustomed to consuming in their daily routine. Feeding amounts were adjusted based on age and body weight to ensure consistency with normal dietary intake patterns. This approach aimed to minimize variability related to feeding habits while maintaining the nutritional requirements of the animals, in line with standard IHUT methodology [11].
Control and test formulas
The same two dry extruded diets were evaluated in the palatability trial: a control diet and a test diet containing MP. The test diet was formulated by including microbial protein at a level of 10% (w/w), hereafter referred to as the MP-containing diet. The control and MP-containing diets were formulated to achieve comparable nutritional targets and differed primarily in their protein source. Minor analytical differences observed after manufacturing reflect normal formulation and analytical variability.
The control diet was formulated to mimic a commercially available complete and balanced formula marketed with a fish and rice flavor profile. Its composition included cereals (12% rice), meat and animal derivatives (34%), poultry fat, krill (3%), beet pulp (1.6%), hydrolyzed animal protein, yeast, carob, minerals, vitamins, and Yucca schidigera (0.001%).
The MP-containing diet was identical to the control diet in overall nutritional composition but differed in ingredient profile by the inclusion of MP, which fully replaced fish meal and partially replaced poultry meal. Due to the inherently low-fat content of microbial protein, poultry fat was adjusted to maintain comparable nutritional targets between diets. The composition of both diets is presented in Table 1, and their analytical constituents determined after manufacturing are reported in Table 2.
Table 1. Ingredient composition of experimental pet food formulations used in palatability assessment.
| Ingredient composition (%) | ||
|---|---|---|
| Control Formula (Fish and Rice) | Test Formula (containing MP) | |
| Cereals | 51% | 51% |
| Poultry Meal | 28% | 22% |
| MP | – | 10% |
| Fish Meal | 3.5% | – |
| Meat and bone Meal | 2% | 2% |
| Poultry fat | 7% | 7.5% |
| Krill | 3% | 3% |
| Beet pulp | 2% | 2% |
| Hydrolyzed animal protein | 2% | 2% |
| Carob | 0.2% | 0.2% |
| Yeast | 0.3% | 0.3% |
| Premix | 0.5% | 0.5% |
| Yucca schidigera | 0.001% | 0.001% |
Table 2. Analytical composition of experimental diets and estimated metabolizable energy.
| Analytical Composition | ||
|---|---|---|
| Control Formula (Fish and Rice) | MP-containing diet | |
| Crude Protein | 30% | 30% |
| Crude Fat | 13% | 13% |
| Crude Fiber | 2.5% | 2.5% |
| Raw Ash | 8.5% | 9% |
| Moisture | 8% | 8% |
| Estimated metabolized energy (kcal/kg) | 3,485 | 3,468 |
Both experimental diets were manufactured under controlled and identical conditions at Ovargado S.A. (São João de Ovar, Portugal) using an industrial-scale single-screw extrusion system (Andritz), with a minimum batch capacity of approximately 3,000 kg. The base mixture was preconditioned with direct steam injection and water addition to facilitate starch gelatinization and initial thermal processing. Extrusion was performed under controlled thermal and mechanical parameters, with barrel temperatures ranging from 120 °C to 140 °C. The extrudate was cut at the die face to form uniform kibbles, which were subsequently dried to a final moisture content below 10%.
Post-extrusion, the dried kibbles were coated using a rotary drum coater operated at ambient temperature and atmospheric pressure. A portion of the total poultry fat (included at 7.0–7.5% of the formulation) was first applied, followed by powdered poultry digest (0.2%, w/w) as a palatability enhancer. The same post-extrusion coating procedure, including identical poultry fat and poultry digest application rates, was used for both the control and MP-containing diets.
Testing locations and product delivery
The experimental trials were conducted in the animals’ habitual feeding environments, predominantly in their domestic settings (Portuguese households). Test products were delivered to participants by Sense Test, Lda. (Vila Nova de Gaia, Portugal) technicians or representatives. During individual meetings, the trial conditions, protocols, and requirements were explained, and participants were provided with written instructions, which they were required to acknowledge before beginning the trial. Pet owners were instructed to feed the diets as part of their pets’ routine feeding practices to ensure consistency. The products were delivered in appropriate containers, each clearly labeled with a code and a generic description of the sample. All samples were coded to maintain blinding and minimize bias throughout the trial. The use of coded labeling ensured that participants remained unaware of the specific formulation of the diets during the study, thereby mitigating any potential subjective influence on feeding behavior and trial outcomes.
Two-bowl preference test
The two-bowl preference test was used to assess short-term food preference and was carried out as follows: each animal was simultaneously presented with two bowls of food, one containing the MP-containing diet and the other containing the control diet [4,8]. The test was conducted over two consecutive days (four meals per animal), representing a short-term assessment of food preference. To minimize potential bias, the positions of the bowls were alternated during each of the four meals. Both bowls were placed in the animals’ feeding area with food portions standardized to identical amounts for both samples, corresponding to the quantities typically provided to the animals in their daily meals. Owners were instructed not to provide treats, table scraps, or any other foods during the two-day study period and to use the feeding bowls provided by Sense Test. The trial was conducted under routine home-feeding conditions to minimize disruption of the animals’ usual feeding behavior.
Data collection and analysis
For each meal, owners used digital scales to record the weight of the empty bowl, the bowl containing food before feeding, and the bowl with the remaining food after feeding. Food consumption was calculated as the difference between the amount of food offered and the amount remaining after each meal. Food consumption data from the four meals were aggregated for each animal prior to statistical analysis, and each animal was considered the experimental unit. Owners also recorded the first bowl approached, the first bowl from which the animal consumed food, and whether the total food intake during the test period was higher, lower, or similar to the animal’s usual intake. These observations were summarized descriptively. Statistical analyses were performed using XLSTAT (Lumivero, USA) and Microsoft Excel. Differences in food intake between the MP-containing and control diets were assessed using the Wilcoxon signed-rank test. Statistical significance was set at p < 0.05. First approach and first consumed were reported descriptively as percentages.
Analytical characterization of experimental diets
Independent laboratory analyses were conducted to characterize control and MP-containing diets and to support interpretation of the palatability results. All analyses were performed by Eurofins Food Testing Lisboa (Lisbon, Portugal) using validated internal methods or internationally recognized standard methods, as specified in the analytical reports.
Proximate composition, including crude fat (after acid hydrolysis) and crude ash, was determined using internal gravimetric methods. Fatty acid profiles were analyzed by gas chromatography with flame ionization detection (GC-FID), with results expressed both as relative percentages and as absolute values (g/100 g) based on fat content.
Amino acid composition was determined after acid hydrolysis according to ISO 13903:2005 using ion chromatography with UV detection (IC-UV). Sulfur-containing amino acids (cysteine and cystine) and methionine were analyzed following oxidative hydrolysis, while tryptophan was quantified separately using liquid chromatography with fluorescence detection (LC-FLD) according to EU Regulation 152/2009.
Selected minerals were quantified using inductively coupled plasma optical emission spectrometry (ICP-OES) following an internal validated method (CON-PV 00006, 2023−12).
To support interpretation of palatability outcomes, a targeted screening of aroma-related volatile compounds was performed using gas chromatography–mass spectrometry (GC-MS) according to an internal laboratory method.
Results
Canine Diets
An IHUT trial was conducted with 30 dogs from different breeds (average body weight 5.9 ± 2.4 kg) to assess short-term food preference of formulations with or without microbial protein. Both diets were formulated to achieve comparable nutritional targets and differed primarily in their protein source. Overall food intake remained similar to the animals’ usual intake, with 91% ± 3% of participants reporting that their dogs consumed the usual amount or more than their regular meals. Only four participants (13%) reported minor alterations in fecal consistency, with two reporting firmer stools and two reporting softer stools.
Regarding food preference, the two-bowl preference test revealed that dogs exhibited a statistically significant preference for the MP-containing diet compared with the control diet. Across the four meals, 58% of dogs first approached the MP-containing diet and 58% consumed it first, compared with 42% for the control diet (Figs 1 and 2). Fig 3 shows that dogs consumed a significantly greater proportion of the MP-containing diet than the control diet (72% vs. 50%; Wilcoxon signed-rank test, p = 0.002).
[图：First approach (Dogs). Percentage of observations in which dogs first approached the MP-containing diet or the control d]
[图：Primary consumption (Dogs). Percentage of observations in which dogs first consumed the MP-containing diet or the contro]
[图：Food intake (Dogs). Mean proportion of the MP-containing diet and control diet consumed by dogs during the two-bowl pref]
Feline Diets
A similar IHUT trial was performed, but this time with 30 cats (average weight 5.3 ± 2.1 kg). Three out of four cats (75% ± 2%) were reported to have maintained or increased their food consumption during the trial compared to their regular meals. Overall, a statistically significant difference in food preference between diets was observed (Wilcoxon signed-rank test, p = 0.001).
On average, across the four meals offered, 62% of cats first oriented toward the bowl containing the MP-containing diet (Fig 4), and 68% consumed this diet first (Fig 5). Mean intake of the MP-containing diet reached 51% of the total offered food, compared with 34% for the control diet (Wilcoxon signed-rank test, p = 0.001; Fig 6).
[图：First approach (Cats). Percentage of observations in which cats first approached the MP-containing diet or the control d]
[图：Primary consumption (Cats). Percentage of observations in which cats first consumed the MP-containing diet or the contro]
[图：Food Intake (Cats). Mean proportion of the MP-containing diet and control diet consumed by cats during the two-bowl pref]
Analytical characterization of control and MP-containing diets
Independent laboratory analyses showed that diets containing microbial protein and the control diets were formulated to achieve comparable nutritional targets rather than identical analytical composition, as shown in Table 3. Minor analytical differences were observed, including ash, tryptophan and the ω6:ω3 ratio, maintaining broadly comparable nutritional profiles. Crude fat was equivalent by analysis (13.1% in both diets), while diets containing microbial protein showed a slightly higher ash content (9.8% vs. 9.0% in the control). Amino acid profiling was broadly comparable between diets. Glutamic acid levels were comparable between diets (3.73 vs. 3.71 g/100 g), whereas tryptophan content was higher in the diets containing microbial protein (0.333 vs. 0.259 g/100 g). Fatty acid profiles were similar overall; however, the omega-6: omega-3 ratio was lower in the diets containing microbial protein (8.3 vs. 11.2 in the control). A targeted volatile compound screen did not identify marked elevations of individual volatile compounds in either diet, with most analytes detected below limits of quantification.
Table 3. Selected analytical characteristics of the control and MP-containing diets.
| Parameter | Control Diet | MP-Containing Diet |
|---|---|---|
| Crude Fat (%) | 13.1 | 13.1 |
| Ash (%) | 9.0 | 9.8 |
| Glutamic Acid (G/100 G) | 3.71 | 3.73 |
| Tryptophan (G/100 G) | 0.259 | 0.333 |
| Ω 6: Ω 3 Ratio | 11.2 | 8.3 |
| Targeted Volatile Screen | Mostly <Loq | Mostly <Loq |
Discussion
In this study, short-term food preference for dry extruded diets containing MP was evaluated using a two-bowl preference test conducted under an In-Home Use Testing (IHUT) protocol, in comparison with a nutritionally equivalent, macronutrient-matched control diet formulated to differ primarily in protein source. The two-bowl test is a widely established methodology in pet food palatability research and is recognized for its ability to provide reliable and reproducible measures of dietary preference [4]. Animals may develop a preference for a specific side of their feeding area due to habit, spatial comfort, or other environmental factors, which could skew the results of the palatability test. By alternating bowl placement, researchers minimize the influence of positional bias,supporting interpretation of the observed preference in relation to the sensory attributes of the food itself, such as taste, aroma, or texture rather than external factors like bowl positioning [4]. Most participants of the study reported a total food intake similar or even higher than the amounts normally consumed by their pet (both cats and dogs). The significance of owner-reported observations in these trials is underscored by research indicating that pet owners can accurately identify and interpret their animals’ feeding behaviors, which correlates strongly with palatability levels measured through quantitative food consumption [12]. An important aspect of the experimental design is that microbial protein completely replaced fish meal, an ingredient generally regarded as highly palatable in companion animal diets. Despite this substitution, both dogs and cats showed a significant preference for the MP-containing diet under the conditions evaluated. These findings indicate that replacement of fish meal with microbial protein did not compromise short-term food preference under the conditions evaluated. The results indicate that the study protocol did not substantially alter the animals’ usual feeding routine, supporting the interpretation of the observed food preference under routine home-feeding conditions. The four-meal evaluation period was selected to reduce the potential influence of an initial novelty response while maintaining the short-term nature of the preference assessment. Although repeated exposure may lessen transient neophilic or neophobic responses, the present study was not designed to characterize these behavioral phenomena independently, and the observed preference should therefore be interpreted within the context of a short-term food preference test. This study’s findings indicate that diets containing MP were preferred over the control diet under the conditions evaluated. These findings support further investigation of microbial protein as an alternative protein source for pet food.
Conclusion
In conclusion, this study indicates that the inclusion of microbial protein at 10% did not impair short-term food preference and was associated with increased preference under the conditions evaluated. Diets containing MP were preferred by both dogs and cats compared with the control diet, providing new evidence of short-term food preference in addition to previously reported nutritional quality and digestibility [6,7]. The inclusion of both canine and feline data represents an important contribution, as palatability studies on alternative protein sources have historically focused primarily on dogs. These findings highlight the potential of MP as a versatile ingredient suitable for multiple companion animal species. Together, these findings support further investigation of MP as an alternative protein source for pet food formulations, alongside its previously reported nutritional quality, consistent availability from controlled production systems [6], and favorable food preference. As this study evaluated only short-term food preference, no conclusions can be drawn regarding long-term feeding behavior, sustained acceptance, or health outcomes. Further research involving longer-term feeding trials across different life stages and breeds is warranted to confirm these findings and expand their applicability in companion animal nutrition.
Supporting information
[内容请查看原文]
[内容请查看原文]
[内容请查看原文]
[内容请查看原文]